INSECTICIDAL ACTIONS OF CITRUS AURANTIUM

INSECTICIDAL ACTIONS OF CITRUS AURANTIUM

ELIAS P. SISKOS

A thesis submitted to Cardiff University in candidature for the degree ofDoctor of Philosophy

Cardiff School of Biosciences Cardiff University

September 2004

UMI Number: U584693

All rights reserved

INFORMATION TO ALL USERS The quality of this reproduction is dependent upon the quality of the copy submitted.

In the unlikely event that the author did not send a complete manuscript and there are missing pages, these will be noted. Also, if material had to be removed,

a note will indicate the deletion.

Dissertation Publishing

UMI U584693Published by ProQuest LLC 2013. Copyright in the Dissertation held by the Author.

Microform Edition © ProQuest LLC.All rights reserved. This work is protected against

unauthorized copying under Title 17, United States Code.

ProQuest LLC 789 East Eisenhower Parkway

P.O. Box 1346 Ann Arbor, Ml 48106-1346

DECLARATION

This work has not previously been accepted in substance for any degree and is not being concurrently submitted in candidature for any degree.

Signed .......................................... (candidate)

Date Z j / . O ?../.&>.?. V........................

STATEMENT 1

This thesis is the result o f my own investigations, except where otherwise stated.

Other sources are acknowledged by footnotes giving explicit references. A bibliography is appended.

Signed ........................................... (candidate)

D ate..............

STATEMENT 2

I hereby give consent for my thesis, if accepted, to be available for photocopying and for inter-library loan, and for the title and summary to be made available to outside organisations.

Signed ..........

Date /.O f),\/.2 &q V

(candidate)

To my Giota

ACKNOWLEDGEMENTS

First of all I would like to thank the person who offered me the theme of the thesis

and his laboratory and made this project possible; my supervisor in Greece, Dr.

Basilios Mazomenos. I am grateful for his continued support all these years, for his

attention and care and inspiring conversation on my work and for correcting the

manuscripts.

I would like to thank my supervisor in Cardiff, Dr. Mark Jervis, for his help, advice,

encouragement and useful suggestions throughout the course of this study and for his

inspiring and constructive criticism on the manuscripts.

Special thanks to Dr. Fragoulis Krokos for introducing me in the chemical aspects of

this study and especially for his invaluable help on spectroscopic analyses.

I would like to thank Dr. David Kelly for providing the NMR facilities and for his

help in the interpretation of the NMR spectra.

Thanks to Prof. May Berenbaum for providing the synthetic osthol.

I would like to thank all the people o f the Chemical Ecology and Natural Products

laboratory, especially Dr. Maria Konstantopoulou and Mrs. Tasoula Pantazi-

Mazomenos, for their support all these years and for providing a friendly and warm

environment.

I should acknowledge the State Scholarship Foundation (Republic of Greece) for the

generous financial support of most of my studies.

Finally, I would like to thank my parents and my brothers for their support and love

during my studies and my life.

SUMMARY

This thesis was aimed at the extraction, isolation and structure elucidation of insecticidally active secondary metabolites of Citrus aurantium plant parts for potential use either as commercial insecticides or as lead compounds.

Fruits, leaves and shoots of C. aurantium were extracted in methanol and the chemicals recovered were fractionated by liquid-liquid partitioning using organic solvents of increasing polarity (petroleum ether, dichloromethane, ethyl acetate). Petri dish exposure bioassays revealed that only the petroleum ether fraction of the fruit methanol extract was toxic against Bactrocera oleae adults. The bioactive chemicals present in the fruits were produced and/or accumulated in the peels.

Petri dish exposure and topical application bioassays revealed that a petroleum ether peel extract was toxic against B. oleae and Ceratitis capitata adults. In several cases, differences in susceptibility were revealed between the two species and between the two sexes of the same species.

First gravity column fractionation of the peel extract revealed that only the F2 fraction was active, while fraction Ft has a synergistic effect. Further purification of the F2 fraction on the second gravity column, followed by HPLC, resulted in the isolation of three major compounds.

The chemical structure of these components was elucidated by spectroscopic methods (UV, FTIR, GC-MS and *H NMR) and was assigned as osthol, bergapten and 6',7'- epoxybergamottin. Of the three isolated compounds, only the 6',7'-epoxybergamottin was active and its toxicity was the same to the synthesised 6',7'-epoxybergamottin. However, the synergism revealed between isolated bergapten and 6',7'- epoxybergamottin was not confirmed when isolated bergapten was replaced by synthetic bergapten, indicating that minor components present in the isolated bergapten were responsible for the synergistic effect.

Crude or semi-purified peel extracts of C. aurantium may have potential for insect control. Moreover, 6',7'-epoxybergamottin has apparently never been reported to have deleterious effects on insects. At this early stage, the potential of 6',7'- epoxybergamottin as a lead compound for a new class of insecticides remains to be determined.

CONTENTS

Acknowledgments i

Summary ii

Contents iii

Abbreviations vi

List of Figures ix

List of Tables xii

1 General Introduction 1

1.1 Insects as Pests 1

1.2 Plant-derived chemicals as a source of insecticides 3

1.3 The family Rutaceae 7

1.3.1 The genus C itrus 1

1.4 The family Tephritidae 12

1.4.1 Bactrocera oleae (Gmelin) (Diptera: Tephritidae) 12

1.4.2 Ceratitis capitata (Wiedemann) (Diptera: Tephritidae) 14

1.5 Aims 18

2 General Materials and Methods 20

2.1 Materials 20

2.1.1 Insects 20

2.1.2 Plant materials 23

2.1.3 Chemicals 25

2.2 Methods 25

2.2.1 Extraction methods 25

2.2.2 Bioassays 26

2.2.3 Statistics 33

3 Screening for insecticidal activity in Citrus aurantium extracts 35

3.1 Introduction 35

3 .2 Methods 40

3 .2.1 Extraction of plant materials 40

iii

3.2.2 Bioassay of fractions and extracts 49

3.2.3 TLC analysis of fractions and extracts 50

3.3 Results 52

3.3.1 Extraction yield, chemical profile and insecticidal activity of various

fractions of CA plant parts 52

3.3.2 Extraction yield, chemical profile and insecticidal activity of

petroleum ether fractions of CA fruit tissues 58

3.3.3 Extraction yield, chemical profile and insecticidal activity of

petroleum ether fractions obtained from fresh and dried peels of CA 62

3.3.4 Effect of different extraction methods and solvents on extraction

yield, chemical profile and insecticidal activity of peel fractions and

extracts of CA 66

3.3.5 Effect of fruit ripeness on extraction yield, chemical profile and

insecticidal activity of petroleum ether peel extracts of CA 70

3.4 Discussion 75

4 Evaluation of the insecticidal activity of a petroleum ether peel extract of

Citrus aurantium to Bactrocera oleae and Ceratitis capitata adults 79

4.1 Introduction 79

4.2 Methods 82

4.2.1 Extraction of CA peels 82

4.2.2 Bioassays 82

4.3 Results 85

4.3.1 Petri dish exposure bioassays 85

4.3.2 Fumigant bioassays 94

4.3.3 Topical application bioassays 97

4.4 Discussion 112

5 Isolation of insecticidally active components from the petroleum ether peel

extract of Citrus aurantium 118

5.1 Introduction 118

5.2 Methods 122

5.2.1 Fractionation procedure of the petroleum ether peel extract of CA 122

5 .2.2 Bioassay of fractions 126

iv

5.2.3 TLC analysis of fractions 127

5.2.4 HPLC analysis of fractions 127

5.3 Results 128

5.3.1 First gravity column fractionation 128

5.3.2 Second gravity column fractionation 134

5.3.3 HPLC fractionation 139

5.4 Discussion 152

6 Identification of isolated insecticidally active components 155

6.1 Introduction 155

6.2 Methods 157

6.2.1 Spectroscopic methods 15 7

6.2.2 Synthesis of 6',7'-epoxybergamottin 159

6.2.3 Bioassay of isolated and synthetic compounds 159

6.3 Results 161

6.3.1 Spectroscopic data of the isolated compounds 161

6.3.2 Synthesised 6',7'-epoxybergamottin 173

6.3.3 Spectroscopic data of synthetic osthol, bergapten and 6',7'-

epoxybergamottin 173

6.3.4 Bioassays 173

6.4 Discussion 178

7 General Discussion 181

8 References 187

v

ABBREVIATIONS

ANOVA: Analysis of variance

BO: Bactrocera oleae

b.p.: Boiling point

br: Broad

°C: Degrees Celsius

ca .: Approximately

CA: Citrus aurantium

CC: Ceratitis capitata

CDC13: Deuteriochloroform

CHCh: Chloroform

cm: Centimeter

Conv. S-L. Conventional solid-liquid extraction method

d: Day or doublet, depending on context

DCI: Direct Contact Inlet

DCM: Dichloromethane

DEPT: Distortionless enhancement through polarisation transfer

df: Degrees of freedom

El: Electron impact (ionisation)

EtOAc: Ethyl acetate

eV: Electron volt

FID: Flame Ionisation Detector

FTIR: Fourier Transform Infrared Spectroscopy

g: Gram

GC: Gas Chromatography

h: Hour

Hex: n-Hexane

H20: Water

HPLC: High Performance Liquid Chromatography

Hz: Hertz

id.: Internal diameter

IPM: Integrated Pest Management

IR:

1:

L C 5 0 :

L D 5 0 :

//A:

/'g 'fiV.

m:

mA:

MCPBA:

MeOH:

mg:

MHz:

min:

ml:

mm:

MS:

m z\

n:

Na2SC>4:nm:

NMR:

P:

PE:

PTFE:

Rr:s:

SE:

sec:

SIT:

t:

Infrared

Litre

Median lethal concentration: Statistically derived concentration

expected to kill 50 % of test organisms in a given population

Median lethal dose: Statistically derived dose expected to kill 50 % of

test organisms in a given population

Microamper

Microgram

Microlitre

Micrometer

Multiplet

Milliamper

3-Chloroperbenzoic acid

Methanol

Milligram

Megahertz

Minute

Millilitre

Millimeter

Mass Spectrometry

Mass-to-charge ratio

Number of sampling units in a sample

Sodium sulphate

Nanometer

Nuclear Magnetic Resonance

Probability

Petroleum ether (b.p. 40-60 °C)

Polytetrafluoroethylene (Teflon)

Retention factor

Singlet

Standard error

Second

Sterile Insect Technique

Triplet

vii

TLC: Thin Layer Chromatography

UV: Ultraviolet

v: Volume

X2: Chi-squared

LIST OF FIGURES

Figure 1.1 Bactrocera oleae, adult female.

Figure 1.2 Ceratitis capitata, adult female.

Figure 2.1 Wood-framed colony cage in which artificially-reared BO adult flies were

maintained.

Figure 2.2 Plexiglas colony cage in which artificially-reared CC adult flies were

maintained.

Figure 2.3 CA trees with fruits at different ripening stages.

Figure 2.4 A Petri dish used in the Petri dish exposure bioassay in which the test sample

was applied to the entire inner surface and thin cardboard strips were placed

between the sides and the lid.

Figure 2.5 Wood-framed cages, containing solid adult diet and water, in which tested

flies were transferred (10 flies/cage) after 12 h exposure to treated and control

Petri dishes.

Figure 2.6 The 4 Petri dishes used in the fumigation bioassay.

Figure 2.7 The Burkard hand micro-applicator, used to apply topically 1 p\ of solvent or

test solution to the dorsal surface of the thorax of each fly.

Figure 3.1 Fruits, leaves and shoots of CA.

Figure 3.2 Partitioning procedure of CA methanol extracts.

Figure 3.3 Fruit tissues of CA.

Figure 3.4 Thin layer chromatogram of various fractions of CA plant parts.

Figure 3.5 Thin layer chromatograms of petroleum ether (A), ethyl acetate (B) and

residual (C) fractions of CA plant parts.

Figure 3.6 Thin layer chromatogram of petroleum ether fractions of CA fruit tissues.

Figure 3.7 Thin layer chromatogram of petroleum ether fractions obtained from fresh

and dried peels of CA.

Figure 3.8 Thin layer chromatogram of peel fractions and extracts of CA obtained by

different extraction methods and initial extraction solvents.

Figure 3.9 Thin layer chromatogram of petroleum ether peel extracts of CA obtained by

fruits at various ripening stages.

Figure 4.1 Percentage mortality of BO male adults exposed to different concentrations

(ug/cm2) of the petroleum ether peel extract of CA using the Petri dish exposure

bioassay.

13

14

22

22

24

28

29

31

32

41

43

45

55

56

60

64

68

72

86

ix

Figure 4.2 Percentage mortality of BO female adults exposed to different concentrations


bioassay. 87

Figure 4.3 Log (ug/cm2) - probit mortality lines calculated by probit analysis for males

and females of BO exposed to the petroleum ether peel extract of CA using the

Petri dish exposure bioassay at 48 (A), 72 (B) and 96 (C) h after treatment. 88

Figure 4.4 Percentage mortality of CC male adults exposed to different concentrations


bioassay. 90

Figure 4.5 Percentage mortality of CC female adults exposed to different concentrations


bioassay. 92

Figure 4.6 Log (ug/cm2) - probit mortality lines calculated by probit analysis for males

and females of CC exposed to the petroleum ether peel extract of CA using the

Petri dish exposure bioassay at 48 (A) and 72 (B) h after treatment. 93

Figure 4.7 Percentage mortality of BO male adults treated with different doses

(ug/insect) of the petroleum ether peel extract of CA using the topical application

bioassay. 99

Figure 4.8 Percentage mortality of BO female adults treated with different doses

(//g/insect) of the petroleum ether peel extract of CA using the topical application

bioassay. 100

Figure 4.9 Log (ug/insect) - probit mortality lines calculated by probit analysis for males

and females of BO treated with the petroleum ether peel extract of CA using the

topical application bioassay at 72 (A), 96 (B), 120 (C) and 144 (D)h after

treatment. 103

Figure 4.10 Log (ug/mg) - probit mortality lines calculated by probit analysis for males

and females of BO treated with the petroleum ether peel extract of CA using the

topical application bioassay at 72 (A), 96 (B), 120(C) and 144 (D) h after

treatment. 104

Figure 4.11 Percentage mortality of CC male adults treated with different doses


bioassay. 106

Figure 4.12 Percentage mortality of CC female adults treated with different doses


bioassay. 107

x

Figure 4.13 Log (//g/insect) - probit mortality lines calculated by probit analysis for

males and females of CC treated with the petroleum ether peel extract of CA

using the topical application bioassay at 24 (A), 48 (B), 72 (C) and 96 (D) h

after treatment. 110

Figure 4.14 Log (/^g/mg) - probit mortality lines calculated by probit analysis for males

and females of CC treated with the petroleum ether peel extract of CA using the

topical application bioassay at 24 (A), 48 (B), 72 (C) and 96 (D) h after

treatment. 111

Figure 5.1 Fractionation procedure of the petroleum ether peel extract of CA. 123

Figure 5.2 Thin layer chromatograms of the fractions obtained from the first gravity

column fractionation of the petroleum ether peel extract. 130

Figure 53 Thin layer chromatograms of the fractions obtained from the second gravity

column fractionation of the F2 fraction. 136

Figure 5.4 HPLC chromatogram of the F22 fraction. 140

Figure 6.1 Chemical structures ofosthol (1), bergapten(2) and 6',7-epoxybergamottin

(3). 161

Figure 6.2 UV spectrum of osthol. 162

Figure 6.3 IR spectrum of osthol. 163

Figure 6.4 MS spectrum of osthol. 164

Figure 6.5 NMR spectrum of osthol. 165

Figure 6.6 UV spectrum of bergapten. 166

Figure 6.7 IR spectrum of bergapten. 167

Figure 6.8 MS spectrum of bergapten. 168

Figure 6.9 'H NMR spectrum of bergapten. 169

Figure 6.10 UV spectrum of 6',7'-epoxybergamottin. 170

Figure 6.11 MS spectrum of 6',7'-epoxybergamottin. 171

Figure 6.12 !H NMR spectrum of 6',7'-epoxybergamottin. 172

Figure 6.13 The chemical structure of bergamottin. 175

xi

LIST OF TABLES

Table 3.1 Collection data for CA samples. 40

Table 3.2 Extraction yield of various fractions of CA plant parts and % contribution of

each fraction to the sum of fractions of the same plant part. 52

Table 3.3 Dry weight (%) of CA plant parts and extraction yield of their sum of fractions. 53

Table 3.4 Percentage mortality of BO adults exposed to various fractions of CA plant

parts at 200 /^g/cm2 using the Petri dish exposure bioassay. 57

Table 3.5 Dry weight (%) of CA fruit tissues and extraction yield of their petroleum

ether fractions. 59

Table 3.6 Percentage mortality of BO adults exposed to petroleum ether fractions of CA

fruit tissues using the Petri dish exposure bioassay. 61

Table 3.7 Percentage mortality of BO adults exposed to petroleum ether fractions

obtained from fresh and dried peels of CA at 100//g/cm2 using the Petri dish

exposure bioassay. 65

Table 3.8 Extraction yield of peel fractions and extracts obtained from 100 g fresh peels

of CA using different extraction methods and initial extraction solvents. 66

Table 3.9 Percentage mortality of BO adults exposed to peel fractions and extracts of CA

obtained by different extraction methods and initial extraction solvents using the

Petri dish exposure bioassay. 69

Table 3.10 Diy weight (%) of peels obtained from CA fruits at various ripening stages

and extraction yield of their petroleum ether extracts. 71

Table 3.11 Percentage mortality of BO adults exposed to petroleum ether peel extracts of

CA obtained by fruits at various ripening stages using the Petri dish exposure

bioassay. 74

Table 4.1 Summary of probit analyses of mortality data for males and females of BO

exposed to different concentrations of the petroleum ether peel extract of CA

using the Petri dish exposure bioassay. 88

Table 4.2 Summary of probit analyses of mortality data for males and females of CC

exposed to different concentrations of the petroleum ether peel extract of CA

using the Petri dish exposure bioassay. 93

Table 4.3 Percentage mortality of BO adults exposed to the petroleum ether peel extract

of CA at 4mg/Petri dish using the fumigation bioassay. 95

Table 4.4 Percentage mortality of CC male and female adults exposed to the petroleum

ether peel extract of CA at 12mg/Petri dish and 24 mg/Petri dish, respectively

using the fumigation bioassay. 96


treated with different doses (//g/insect) of the petroleum ether peel extract of CA

using the topical application bioassay. 101


treated with different doses (ug/mg) of the petroleum ether peel extract of CA



treated with different doses (wg/insect) of the petroleum ether peel extract of

CA using the topical application bioassay. 108


treated with different doses (wg/mg) of the petroleum ether peel extract of CA


Table 5.1 Weights of fractions and their % recovery after the first gravity column

fractionation of 2 g of the petroleum ether peel extract of CA. 128


of CA and its fractions obtained from the first gravity column fractionation using

the Petri dish exposure bioassay. 131


of CA and the sum of its fractions obtained from the first gravity column

fractionation using the Petri dish exposure bioassay. 133


of CA and different binary blends of its fractions obtained from the first gravity

column fractionation using the Petri dish exposure bioassay. 134

Table 5.5 Weights of fractions and their % recovery after the second gravity column

fractionation of 1 g of the F2 fraction. 135

Table 5.6 Percentage mortality of BO adults exposed to the F2 fraction and its fractions

obtained from the second gravity column fractionation using the Petri dish


Table 5.7 Percentage mortality of BO adults exposed to the F2 and F22 fractions using the


Table 5.8 Weights of fractions and their % recovery after the HPLC fractionation of 100

mg of the F22 fraction. 141

Table 5.9 Distribution (%) of the three major peaks of fraction F22 to the fractions

obtained from the HPLC fractionation of 100 mg of the F22 fraction. 142

Table 5.10 Percentage mortality of BO adults exposed to the F22 fraction and its fractions

obtained from the HPLC fractionation using the Petri dish exposure bioassay. 143

Table 5.11 Percentage mortality of BO adults exposed to the F22 and F226 fractions using

the Petri dish exposure bioassay. 144

Table 5.12 Percentage mortality of BO adults exposed to the F22 fraction and the sum of

fractions obtained from the HPLC fractionation using the Petri dish exposure

bioassay. 146

Table 5.13 Percentage mortality of BO adults exposed to the F22 fraction and different

combinations of the fractions obtained from the HPLC fractionation using the



binary blends of the three major peaks of the F22 fraction using the Petri dish



of CA and a blend consisted of the Fi, F224 and F226 fractions using the Petri dish


Table 6.1 Summary of probit analyses of mortality data of BO exposed to different

concentrations of the isolated and the synthetic 6',7'-epoxybergamottin and the

botanical insecticide rotenone using the Petri dish exposure bioassay. 174

Table 6.2 Percentage mortality of BO adults exposed to different combinations of

isolated and synthetic 6',7'-epoxybergamottin and bergapten using the Petri dish


xiv

CHAPTER 1

GENERAL INTRODUCTION

1.1 INSECTS AS PESTS

From a purely anthropocentric standpoint, insects are pests when they harm man, his

crops, animals or property. More specifically, in agronomy an insect is classified as a

pest if the damage it causes to a crop or to livestock is sufficient to reduce the yield

and/or quality of the harvested product by an amount that is unacceptable to the

farmer (Dent, 1991). The estimated number of insect species varies from less than 5

million to as many as 80 million (Gullan and Cranston, 1994). However, only around

1 million insect species have been described up to now. O f these, fewer than 10,000

can be considered as pests, some 3,500 requiring regular attention and as few as 600

requiring control measures (Pedigo, 1996).

Insects may become crop pests for many reasons:

■ The accidental or intentional introduction of insects into a favourable new area

without the controlling influence of their natural enemies (Gullan and

Cranston, 1994).

■ The adoption of introduced crop species by native insects (van Driesche and

Bellows, 1996).

■ The simplification of agroecosystems by large-scale culture of a single crop,

with uniform planting and harvesting schedules (van Driesche and Bellows,

1996).

1

■ The use of broad spectrum insecticides which cause non-pest species to

become major pests by eliminating their natural enemies (Johansen, 1962).

■ The use of less resistant, but more high-yielding cultivars compared to wild

relatives (Dent, 1991).

■ The adoption of cultural practices, such as the continued cultivation of a single

crop, without crop rotation, sanitation and a fallow period, which allows the

build-up of insect pest numbers (Gullan and Cranston, 1994).

Since the development of agriculture, man has employed a variety of control measures

to protect his crops from pest insects. These are: control through plant resistance

(Dent, 1991), cultural control (Coaker, 1987), biological control (van Driesche and

Bellows, 1996), chemical control (Matsumura, 1975) and control techniques that

interfere with the normal physiological function or behaviour of insect pests (Dent,

1995) such as the use of pheromones (Jones, 1998) and the sterile insect technique

(Boiler, 1987). However, since the development of DDT the control of insect pests

has been accomplished largely by the use of synthetic insecticides.

Over the past 30 years the concept of Integrated Pest Management (IPM) has

completely changed the philosophy of pest control. Instead of using a single control

technique (mainly the application of pesticides), to control a specific pest problem,

various control measures are combined to control the whole range of pest species in a

cropping system. IPM is a pest management strategy that, in the socioeconomic

context of farming systems, the associated environment and the population dynamics

of pest species, utilises all suitable techniques and methods in as compatible a manner

2

as possible, and maintains the pest population levels below those causing economic

injury (Dent, 1991).

Plant-derived chemicals can play a critical role in IPM systems, as their properties -

rapidly biodegradable, species-specific, environment-friendly - are compatible with

the demands of IPM. However, it is generally acknowledged that synthetic

insecticides are more effective, efficient and stable than natural insecticides (Assabgui

et al., 1997).

1.2 PLANT-DERIVED CHEMICALS AS A SOURCE OF INSECTICIDES

Plant-derived chemicals have played a crucial role in the history of humankind, as

people had recognised and used plant-derived substances to cure illness and to protect

their crops from pests (Hedin and Hollingworth, 1997). The estimated number of

plant species in the world is 0.4-0.5 million (Hostettmann and Wolfender, 1997).

However, only a tiny fraction of these species (around 10 %) has been investigated

phytochemically and an even smaller fraction has been screened for activity against

insect pests (Benner, 1993). Furthermore, even plant extracts that have been judged to

be not toxic against a specific insect pest can exhibit considerable toxicity when tested

against other insect pests, as secondary plant metabolites are often active against a

narrow spectrum of insects (Ahn et al., 1997). Thus, it can be concluded that there is a

great potential for isolating chemicals from plants that can exhibit a range of

biological activity against insects pests. These chemicals can cause acute toxicity, can

be antifeedants, oviposition deterrents and/or repellents, and/or can interfere with

growth and/or development (Coats, 1994).

3

Plants are known to produce a variety of secondary metabolites such as terpenoids,

alkaloids, phenolics, polyacetylenes, flavonoids, unusual amino acids and sugars (Ahn

et al., 1997; Hostettmann and Wolfender, 1997). These substances have no apparent

function to primary metabolism and their main role is to protect plants from microbial

pathogens, insects, vertebrate herbivores and other plants (Coats, 1994). Other roles

they have are as pollinator attractants and as chemical adaptations to environmental

stresses (Balandrin et al., 1985).

The renewed interest over recent years in plant-derived chemicals as insecticides has

been fuelled by .

■ The increasing political and consumer pressures to reduce the usage of

synthetic insecticides due to their adverse environmental effects, residues on

agricultural products and uncertain long-term ecological and biological effects

(Isman et al., 1997).

■ The decreasing efficacy of synthetic insecticides due to the development of

resistance (Ahn et al., 1997).

■ The decreasing developmental rate of new synthetic insecticides, due to

stricter requirements on efficacy, selectivity, toxicology and general

environmental impact (McLaren, 1986).

■ The lower costs of the discovery and development of natural insecticides

compared to synthetic insecticides (McChesney, 1994).

■ The development of more sophisticated techniques and instruments for

isolating, purifying and characterising the active substances from plant

extracts (Hedin et al., 1997).

4

Plant-derived insecticides appear to have several advantages over conventional broad-

spectrum insecticides:

■ They are likely to be more rapidly degradable, and consequently have lower

environmental persistence than conventional insecticides; thus, they pose

much less of a threat to the environment (Nair, 1994).

■ They may also show increased specificity and so reduce the risks to beneficial

organisms and consequently reduce the likelihood of outbreaks of secondary

pests (Plimmer, 1993).

■ In the case that new modes of action are validated, insecticide resistance

problems may be reduced (Plimmer, 1993).

Chemical substances isolated from plants for their insecticidal properties can be used

in three different ways to combat insect pests:

■ In very exceptional cases as products per se (Benner, 1993).

■ As lead structures for programmes of synthetic chemistry. The success story of

the synthetic pyrethroids is the best example supporting the use of natural

products as leads for further chemical synthesis (Miyakado et al., 1997).

■ As a source of new modes of action (Escoubas et al., 1994), which is very

important as the major compounds currently used to control insect pests are

neurotoxins acting at limited targets (Copping and Hewitt, 1998).

There are numerous examples in the literature of plant-derived chemicals with

insecticidal properties. The most important and significant application of a plant

natural product centres on the insecticidal properties of pyrethrum, which is obtained

from Chrysanthemum cinerariaefolium (Asteraceae). Pyrethrum is a powerful

5

insecticide which causes a rapid paralysis (knockdown), but it is unstable to sunlight

and rapidly hydrolysed. Pyrethrum, which is a contact nerve poison, has served as the

model for synthesis of the synthetic pyrethroids, several of which are of considerable

current agricultural importance (Graves et al., 1999).

The botanical insecticide which is attracting considerable interest nowadays is neem,

a derivative of the seeds of the Indian neem tree, Azadirachta indica (Meliaceae).

Neem is unusual among botanicals in that it is not an acute toxin. Its numerous

reported effects include repellency, feeding deterrency, oviposition deterrency,

interference with growth and development, and reproduction (Schmutterer, 1990).

A very well-known compound which has been used as a commercial insecticide is

nicotine. Nicotine is found in many species of Nicotiana (Solanaceae). Nicotiana

rustica is a much better source of this compound than the more familiar Nicotiana

tabacum and has been cultivated specifically for nicotine extraction (Benner, 1993).

Rotenone is a botanical insecticide which still retains popularity with gardeners.

Rotenone and related rotenoids are derived from the roots and tubers of South

American and Southern Asian legumes in the genera Derris, Lonchocarpus and

Tephrosia (Benner, 1993). Other botanical insecticides include sabadilla, ryania,

hellebore, strychnine, nomicotine, and anabasine (Graves et al., 1999).

Jacobson (1989) pointed out that the most promising botanicals as sources of novel

plant-based insecticides for use in the present and in the future are species of the

6

families Meliaceae, Rutaceae, Asteraceae, Anoonaceae, Labiatae and Canellaceae.

Among these six plant families, the Meliaceae and the Rutaceae predominate.

1.3 THE FAMILY RUTACEAE

The Rutaceae is distributed in warm and tropical areas of both hemispheres, with its

centres of speciation in South Africa and Australia. There are 150 genera and some

900 species. The most important genus commercially is Citrus.

1.3.1 The genus Citrus

Citrus species are small, often spiny shrubs and trees of the tropics and subtropics

native to Asia and Malaysia. A number are cultivated for their edible fruit: citron,

orange, sweet orange, lemon, lime, pomelo, grapefruit, mandarin orange and

tangerine.

Citrus plants have long been known to be a source of insect control agents. Back and

Pemberton (1915) reported that Citrus peel oils were toxic to eggs and larvae of the

Mediterranean fruit fly, Ceratitis capitata. Volatiles from Citrus peel oils have also

been found to be toxic to eggs and larvae of the Caribbean fruit fly, Anastrepha

suspense (Greany et al., 1983). Abbassy et al. (1979) studied the contact activity of

grapefruit, lemon, lime, mandarin, navel orange, native orange and sweet orange peel

oils against two stored-product insects by the Petri dish exposure bioassay. They

found that against the confused flour beetle, Tribolium confusum, lime oil was the

most toxic, followed by navel orange, lemon, grapefruit and mandarin oils; sweet and

native orange oils were the least toxic. Against granary weevil, Sitophilus granarius,

lime oil was again the most toxic, followed by the oils of grapefruit, navel orange and

7

mandarin; lemon and sweet orange oils were the least toxic, while native orange oil

was not toxic.

Sheppard (1984) demonstrated both the fumigant activity of grapefruit, lime, lemon

and orange peel oils against foraging workers of the red imported fire ant, Solenopsis

invicta, and the contact toxicity of orange peel oil by the topical application bioassay

against house flies, stable flies, black soldier flies, paper wasps and house crickets.

More recently, Ezeonu et al. (2001) studied the fumigant activity of the volatile

extracts of sweet orange and lime against mosquito, cockroach and house fly. These

authors found that volatiles of sweet orange showed greater insecticidal potency,

while the cockroach was the most susceptible to the orange peels among the three

insects studied.

Limonene, a major constituent of Citrus oils, exhibited contact and fumigant activity

against several insects (Taylor and Vickery, 1974). Rani and Osmani (1980) reported

that citral, a minor constituent of Citrus oils, was very active as a fumigant against

house fly; contact toxicity was also exhibited by citral but to a lesser degree. Styer and

Greany (1983) compared the insecticidal activity of citral and limonene against the

Caribbean fruit fly. They found that citral was considerably more toxic than limonene

to the eggs and larvae of this species. However, citral and limonene were equally

effective to house fly adults by the topical application and moreover they were toxic

to adults of the twospotted spider mite, Tetranychus urticae, by the leaf-dip bioassay

(Lee et al., 1997).

8

Su et al. (1972b) demonstrated that even when the volatile materials (about 30-35 %,

mostly terpene compounds), were removed by lyophilisation from Citrus peel oils, the

remaining non-volatile components exhibited activity when applied topically to

stored-product insects. Terpeneless peel oils of lemon, grapefruit, lime, kumquat,

tangerine, orange, tangelo and temple orange were moderately toxic to rice weevil at

dosages of 25 and 50 //g/cm2. Moreover, terpeneless peel oils of lemon, grapefruit,

lime, kumquat and tangerine were highly toxic to cowpea weevil, whereas orange,

tangelo and temple orange oils were inactive. Furthermore, none of the 8 peel oils

studied showed any appreciable toxicity to black carpet beetle larvae, Indian meal

moth larvae, cigarette beetle, red flour beetle and confused flour beetle. In addition,

Su et al. (1972a) examined the efficacy of these 8 oils as surface protectants of black

eyed peas from cowpea weevil attack. They reported that all oils tested were effective

in preventing nearly all development of an Fj generation at 1.0 and 0.75 % by weight

of the peas.

Salvatore et al. (2004) showed that diethyl ether, ethyl acetate and methanol lemon

peel extracts were toxic against C. capitata larvae when these extracts were

incorporated into their diet. Moreover, a crude aqueous extract of lemon peel was

toxic to Culex pipiens larvae (Thomas and Callaghan, 1999). However, even

powdered sun-dried Citrus peels possess insecticidal chemicals: Don-Pedro (1985)

demonstrated that powdered sun-dried orange and grapefruit peels applied as food

admixtures, exhibited toxic and repellent properties against Dermestes maculatus and

Callosobruchus maculatus.

9

Limonoids, a group of chemically related bitter tetranortriterpene derivatives, are the

most distinctive secondary metabolites of the plant order Rutales (Champagne et al.,

1992). Citrus limonoids, present in large quantities in the by-products of Citrus

industry (Klocke and Kubo, 1982), may act as either feeding deterrents or toxins,

depending on the insect species (Mendel et al., 1991b). The three major neutral

limonoids occurring in the seeds of Citrus species are limonin, nomilin and

obacunone (Mendel et al., 1991b). Limonin, the principle bitter component of Citrus,

represents a potential pest management tool for many insect species (Klocke and

Kubo, 1982). Its antifeedant activity has been reported against the Colorado potato

beetle, Leptinotarsa decemlineata (Alford et al., 1987; Mendel et al., 1991a, 1991b)

and the fall armyworm, Spodoptera frugiperda (Mendel et al., 1991a, 1993). Nomilin

has been found to deter feeding of Reticulitermes speratus (Serit et al., 1992),

Ostrinia nubilalis (Amason et al., 1987), Leptinotarsa decemlineata (Mendel et al.,

1991b) and Spodoptera frugiperda (Mendel et al., 1993). Antifeedant activity has also

been reported for obacunone towards Reticulitermes speratus (Sent et al., 1992) and

Leptinotarsa decemlineata (Mendel et al., 1991b). Limonin, obacunone and nomilin

were reported to act not only as feeding deterrents against the Colorado potato beetle,

but also as toxins (Mendel et al., 1991b). Moreover, these three limonoids were

reported to exhibit moult inhibiting activity in mosquito Culex quinquefasciatus

larvae (Jayaprakasha et al., 1997).

Apart from the insecticidal properties of Citrus species, many other activities of these

species have been reported. Volatile components of several Citrus fruit essential oils

reported to exhibit antimicrobial action on Penicillium digitatum and Penicillium

italicum (Caccioni et al., 1998). Citrus peel and seed extracts have an interesting

10

antioxidant activity with regard to citronellal (Bocco et al., 1998). Nogata et al.

(1996) reported that Citrus fruit extracts inhibited the enzyme activity of rat platelet

cyclooxygenase and lipoxygenase. Citrus aurantium fruit extracts reduced food intake

and body weight gain, in rats and also produce significant pathological changes in

electrical activity of myocardium; their potential as antiobesity herbal medicine was

discussed (Calapai et al., 1999). Moreover, the antimicrobial, antifeeding, insecticidal

and antitumour activities of Citrus paradisi are well documented (Tirillini, 2000).

Although the insecticidal properties of Citrus species are well documented, bitter

orange C. aurantium, has received little attention, like many other members of this

genus. Moreover, most reports concerning studies on the insecticidal properties of

Citrus species were focused on peel and seed extracts and not enough information is

available on the insecticidal properties of extracts derived from other Citrus plant

parts (e.g. leaves, shoots, etc.). In this study C. aurantium has been chosen in order to

study in detail the insecticidal potency o f some of its tissues.

Citrus aurantium (Linnaeus) (Rutaceae)

Citrus aurantium, also known as bitter or sour orange, is a small tree with a smooth,

greyish-brown bark and branches that spread into a fairly regular hemisphere. The

leaves are evergreen, 3 to 4 inches long, glossy, dark green on the upper side, paler

beneath and have sometimes a spine in the axil. The bitter orange and edible orange

trees closely resemble one another, but their leaf-stalks show a marked difference: that

of bitter orange being broadened out in the shape of a heart. The fruit is globular, a

little rougher and darker than that of the common sweet orange.

11

1.4 THE FAMILY TEPHRITIDAE

The Tephritidae (fruit flies) is a moderately large family with about 4500 species

which belongs to the order Diptera (flies). Fruit flies are distributed throughout the

temperate, subtropical, and tropical areas of the world. Individual flies vary in body

length from 1 to over 20 mm. Adult flies are often brightly coloured, their wings may

be banded, they feed on sugary food and they are long-lived (5-6 months). The

females of the main agricultural pests lay eggs in fruits, the developing larvae feed

inside the fruit and most leave when fully grown to pupate in soil. These pests are

thus difficult to kill as larvae, and most control measures have to be applied against

the adult flies. Pest species of Tephritidae fall mostly into the following genera:

Anastrepha, Ceratitis, Bactrocera and Rhagoletis (Hill, 1997).

1.4.1 Bactrocera oleae (Gmelin) (Diptera: Tephritidae)

Bactrocera oleae, also known as the olive fruit fly, exists in all Mediterranean olive-

growing countries and is considered to be the most serious pest of olive fruits.

The adult (Figure 1.1) measure 4 to 5 mm in length. The thorax is reddish-yellow,

with a black dorsum encircled by four greyish bands. The wings are hyaline, with

veins and a dark mark at the apex. The abdomen is of blond colour with segments one

to four decorated with two lateral black marks of varying size. Larvae are white when

living in green olives and dirty-looking purple when living in black olives, and are ca.

7 mm long when full-grown. The colour of the puparium varies from pale yellow to

brown.

12

Figure 1.1 Bactrocera oleae, adult female.

B. oleae's oviposition and food acquisition for larval growth are restricted to fruits of

the genus Olea, from cultivated as well as wild species. Other essential activities and

resources, such as adult feeding, mating and shelter sites, may also be found on non

host plants. The B. oleae female is attracted to the host plant when the olives are

suitable for oviposition. Each female tends to oviposit on suitable olives where no

other egg had previously been laid. The annual population development of B. oleae is

interrupted in northern Mediterranean regions mostly by the harsh climatic conditions

of winter. In intermediate Mediterranean zones, it is interrupted by unfavourable

conditions occurring both winter and summer. In southern Mediterranean regions, B.

oleae population development is interrupted only by the high temperature-low relative

humidity conditions of summer.

Attack by B. oleae may potentially account for 50-60 % of the total insect pest

damage and falls into three main categories: (a) premature fruit fall, (b) decreased

13

yield and quality of oil and (c) spoiling of fruit for consumption as table olives

(Walton, 1995).

1.4.2 Ceratitis capitata (Wiedemann) (Diptera: Tephritidae)

Ceratitis capitata, also known as the Mediterranean fruit fly or Med fly, is the single

most important pest of the family Tephritidae. It is native to central Africa and as a

result of transport by humans, it is now found in practically all the subtropical regions

of the world.

The adults (Figure 1.2) are similar in size to house flies, have a yellow and black

thorax and a yellow abdomen with two silver crossbands, while their wings are

transparent with bands of yellow, brown, and black. Larvae are white, ca. 10 mm long

when full-grown, and the puparium is brown.

Figure 1.2 Ceratitis capitata, adult female.

14

Adults or puparia overwinter in cool regions, but in warmer areas, the species is active

all year. Female flies, which may produce up to 600 eggs, oviposit 2-10 eggs in holes

under the skin of the fruit. Eggs hatch in 2-20 days, and larvae feed within fruits for

about one week. After completion of development, larvae fall to the ground and

pupate in soil for ca. 10 days. Generation time is variable (3-12 weeks), depending on

climatic conditions.

The Med fly is among the most destructive pests of more than 100 species of fruit

including orange, peach, grapefruit, plum, apple, and pear. In subtropical areas, larvae

feed on the pulp of the fruits, causing the development of infection courts for fruit

diseases. Also, holes made by female flies during oviposition scar the fruits, which

lowers quality (Pedigo, 1996).

The predominant method of control of B. oleae and C. capitata during recent decades

has been through the use of conventional pesticides. Two types of insecticide

treatment have generally been used: (a) cover sprays with organophosphorous

insecticides which kill adults (on contact) and larvae in the fruit (by a semi-systemic

effect) and (b) bait sprays using protein hydrolysates and insecticide which attract and

kill the adults By applying bait sprays, only a part of the tree need to be sprayed and

as a result the quantity of insecticide required, the damage to the fauna of beneficial

insects as well as the insecticide residues, is greatly reduced (Katsoyannos, 1992).

However, ecological, toxicological and environmental problems related to the use of

pesticides have given the impetus for the development of alternative methods of

controlling these pests. These are: the use of traps, the sterile insect technique (SIT),

biological control (Kapatos et al., 1977; Wong et al., 1992), insect growth regulators

15

(Budia and Vinuela, 1996; Mazomenos et al., 1997), microbial insecticides (Alberola

et al., 1999; Peck and Mcquate, 2000) and botanical insecticides (Stavroulakis et al.,

2001 ).

A variety of traps utilising one or more attractants (visual, food and sex) have been

designed and evaluated and several attempts to control these flies with such traps have

been made, with encouraging results (Broumas and Haniotakis, 1994; Katsoyannos

and Papadopoulos, 2004). Whether these results can be considered satisfactory for

control purposes, depends on the population levels of the species in the area, the

density of the traps used and the effectiveness of the traps during a particular period

(Kapatos, 1989).

Yellow sticky traps, baited with a food attractant (usually slow release ammonium

salts) and the female sex pheromone (Baker et al., 1980; Mazomenos and Haniotakis,

1981, 1985) for B. oleae and different parapheromones (trimedlure, methyleugenol

and 'Cue-lure’) (Jones, 1998) for C. capitata give very acceptable crop protection at

low to medium pest population densities (Bueno and Jones, 2002; Katsoyannos and

Papadopoulos, 2004).

One disadvantage of such trap, which is based totally or partially on attraction to

colour, is that it also attracts non-target insects, and at high trap densities can cause

damage to beneficial insect populations (Neuenschwander, 1982). By using grey,

green or brown colours attraction of beneficial insects is reduced and thus their

populations are conserved (Mazomenos et al., 2002).

16

Sticky traps used for mass trapping suffered from their limited catching capacity

(especially in areas with high populations). Sticky traps have thus been superseded by

'target’ devices treated with insecticides to kill flies once they have made contact with

the device. Using such ‘lure and kill’, problems of trap saturation are overcome

because the insect, having picked up a lethal dose of insecticide from the target

device, then flies or walk away from it until the toxic effects of the insecticide

manifest themselves (Jones, 1998).

An alternative method to control C. capitata is the sterile insect technique. The SIT is

amongst the most non-disruptive pest control methods. Unlike some other

biologically-based methods it is species-specific, does not release exotic agents into

new environments and does not even introduce new genetic material into existing

populations as the released organisms are not self-replicating. However, the SIT is

only effective when integrated on an area-wide basis, addressing the total population

of the pest, irrespective of its distribution (Hendrichs et al., 2002). In different parts of

the world (e.g., North, Central and South America, Japan, Australia, etc.) the massive

releases of sterile flies proved to be a technology capable of suppressing or

eradicating C. capitata populations on an area-wide scale with negligible adverse

effects to the environment (Schwarz et al., 1989; Hendrichs, 1996). In recent years

this control method has become even more cost-effective due to new technological

breakthroughs such as better diets for mass-rearing, development of male only strains,

increased precision in sterile fly releases, better sterile: fertile discrimination

techniques and more sensitive monitoring networks (Hendrichs et al., 1995;

Hendrichs et al., 2002).

17

SIT has been examined for use against B. oleae and much knowledge has been gained

on rearing techniques and the quality and competitiveness of the sterile flies.

However, the two small scale field trials made in Greece in 1973 and 1974 and again

in 1979, 1980 and 1981, were only partially successful and there are still a number of

political and techno-biological problems that have to be adequately resolved before

SIT could be used for the control of B. oleae (Kapatos, 1989). Several differences in

physiological and behavioural characteristics between mass-reared and wild flies have

been reported, which may account for the failure of SIT control. These include

longevity, reproductive pattern and capacity, male competitiveness, flight ability, field

dispersal, eye colour and vision and pheromone production (Katsoyannos, 1992).

1.5 AIM AND OBJECTIVES

The aim of this study is to investigate whether C. aurantium plant parts synthesise

and/or accumulate chemical compounds that can be used as alternatives of the

synthetic insecticides currently used for the control of Tephritidae.

The objectives of this study are:

■ To screen C. aurantium fractions and extracts for insecticidal activity

against B. oleae adults.

■ To investigate the insecticidal activity of the most promising C. aurantium

extract by contact and fumigant bioassays to both B. oleae and C. capitata

adults.

■ To isolate the active substances of the most promising C. aurantium

extract by using bioassay-guided procedures and B. oleae as the test insect.

18

■ To elucidate the structure of the isolated substances by spectroscopic

methods and to compare the activity of the isolated substances with

synthetic ones.

19

CHAPTER 2

GENERAL MATERIALS AND METHODS

Methods specific to individual chapters are described in those chapters.

2.1 MATERIALS

2.1.1 Insects

Artificially-reared Bactrocera oleae (BO) flies, were mainly used in this study, while

artificially-reared Ceratitis capitata (CC) flies, were used only in order to compare the

susceptibility of the two insects to the final petroleum ether peel extract of Citrus

aurantium (CA). Preliminary studies showed that, in both fly species, there were no

differences in susceptibility between artificially-reared and wild adults. Laboratory-

reared flies were preferred, because they can be produced in high numbers under

standard conditions, their development is faster, and they are available over the whole

year. Wild insects, by contrast, are available only 3-4 months per year, and their rate

of development depends upon environmental conditions which vary significantly.

Laboratory’-reared flies

Adults of both fly species were obtained from artificially-reared colonies maintained

in the insectarium of the Biology Institute NCSR Demokritos for many generations.

The rearing method for BO under laboratory conditions was described by Tsitsipis

(1975, 1977a, 1977b). CC was reared under laboratory conditions using the method

used for BO, with slight modifications concerning caging and oviposition system. The

same adult and larval diets were used for both insect species. All the insects’ stages

20

were maintained in separate rooms under the same rearing conditions: 25 ± 2 °C, 65 ±

5 % relative humidity, 12 h: 12 h light:dark regime and artificial light (3000 lux)

provided by fluorescent light. BO and CC adults were maintained in wood-framed

cages (100 cm long x 40 cm wide x 30 cm high) (Figure 2.1) and in Plexiglas cages

(30 x 30 x 30 cm) (Figure 2.2) respectively and provided with adult solid diet and

water. Adult solid diet was a mixture of yeast hydrolyzate (Nutritional Biochemical

Corporation, USA), sucrose and fresh egg yolk, in a ratio of 1:4:0 7, plus 0.05 %

streptomycin sulphate (Tsitsipis, 1977b). After collection, eggs were transferred with

a brush onto a filter paper impregnated with 0.3 % propionic acid, in a Petri dish, for

an incubation period of 24 h at 25 °C (Manoukas and Mazomenos, 1977) and

subsequently placed on the larval rearing medium. Around 5 eggs were placed for

each gram of larval diet (Tsiropoulos and Manoukas, 1977). The latter had the

following composition (Tsitsipis, 1977b): tap water 55 ml, cellulose powder (No. 123,

Schleicher and Schiill, Germany) 30 g, brewer’s yeast (Schwechat, Austria) 7.5 g,

soya hydrolyzate enzymatic (Nutritional Biochemical Corporation) 3 g, sucrose 2 g,

olive oil (0-1 acidity) 2 ml, Tween-80 0.75 ml, potassium sorbate (Merck, Germany)

0.05 g, Nipagin (methyl-p-hydroxybenzoate) (Merck) 0.2 g, and HC1 solution (2N) 3

ml. The fully developed larvae were allowed to pupariate in sawdust. The puparia

were sieved to remove the sawdust 1-2 days before adult emergence and then placed

in open plastic boxes which were finally introduced into the colony cages.

21

Figure 2.1 Wood-framed colony cage in which artificially-reared BO adult flies were

maintained.

Figure 2.2 Plexiglas colony cage in which artificially-reared CC adult flies were

maintained.

22

m f*

Handling o f flies used in experiments

2-3 d-old adults were used in all bioassays. To ensure that adult flies used in this study

were approximately of the same age, they were allowed to emerge for 24 h in a

Plexiglas cage (30 x 30 x 30 cm) and then the remaining puparia were transferred

every 24 hours to separate cages. Sufficient number of insects of the same generation

(batch) for the five successive days of screening experiments was achieved, by

placing half of the puparia at the optimum storage temperature (25 °C) and the other

half at 18 °C. When all puparia are placed at 25 °C, eclosion of flies occurs for only 3

days. By placing half of the puparia at 25 °C and half at 18 °C, eclosion of flies occurs

for 5 days, as a result of delayed eclosion of flies placed at 18 °C (Tsiropoulos, 1972).

2.1.2 Plant materials

All plant materials used in this study were collected from CA trees (Figure 2.3)

located in the Zografos area of Athens, Greece. Ten trees were selected for sampling.

These were approximately 10 years old, of similar size (4-5 m high) and healthy. In

each collection the same numbers of fruits, leaves and shoots were obtained from each

of the 10 sampling trees. The plant samples were transferred immediately to the

laboratory and cleaned thoroughly in the following order: washed in detergent, rinsed

with tap water and finally rinsed with distilled water. All plant materials were

extracted fresh unless otherwise stated.

23

K>4-

Figure 2.3 CA trees with fruits at different ripening stages.

(A) Unripe fruits (green); (B) Semi-ripe fruits (green-yellow); (C) Semi-ripe fruits (yellow-orange); (D) Ripe fruits (orange).

2.1.3 Chemicals

All chemicals were stored according to supplier’s recommendations.

Standards

Bergamottin: Rotichrom HPLC, Carl Roth, Germany.

Bergapten: 99 %, Aldrich, USA.

Rotenone: 95-98 %, Sigma, USA.

Reagents

3-Chloroperbenzoic acid (MCPBA): ~ 70 %, Fluka, Switzerland.

Sodium sulphate (Na^SOA: Sodium sulphate anhydrous GR, Merck, Germany.

Solvents

Dichloromethane. ethyl acetate, n-hexane, methanol, petroleum ether (b.p. 40-60 %)

and water, were HPLC grade (Lab-Scan, Ireland).

2.2 METHODS

2.2.1 Extraction methods

Two extraction methods were employed throughout this study: conventional solid-

liquid extraction and ultrasound extraction. Prior to extraction, plant materials were

homogenised with the appropriate extraction solvent (approximately 1 ml/g fresh

plant material or dried equivalent to 1 g fresh) for 5 min in a blender.

25

Conventional solid-liquid extraction

Homogenised plant materials (100 g fresh or dried equivalent to 100 g fresh) were

placed in a 2 1 Erlenmeyer flask and cold solvent (20-22 °C) was added (final

extraction solvent volume 1 1). The flask was then placed on a shaker (Orbital Shaker

SOI, Stuart Scientific, UK). Conventional solid-liquid extraction involved the

maceration of the plant material with the extraction solvent for 24 h at room

temperature (20-22 °C). Continuous shaking was applied in order to increase the

efficiency of the method (Silva et al., 1998). Although the solubility of compounds in

a solvent increases with increasing solvent temperature, and also higher temperatures

facilitate penetration of the solvent into the cellular structures of plant materials

extracted, hot extraction solvents were not used due to the possibility that some

compounds could be unstable at higher temperatures, giving rise to compound

artifacts (Houghton and Raman, 1998).

Ultrasound extraction

Homogenised plant materials (100 g fresh) were placed in a 2 1 Erlenmeyer flask and

cold solvent was added (final extraction solvent volume 300 ml). The ultrasound

extraction method involved the sonication of the samples in an ultrasonic bath

(Bransonic Ultrasonic Cleaner B-5200 E l, Branson Ultrasonics Corporation, USA) at

room temperature for a total of 30 min.

2.2.2 Bioassays

Three types of bioassay were used for the evaluation of the insecticidal activity of CA

fractions and extracts: (a) Petri dish exposure bioassay, (b) Fumigation bioassay and

(c) Topical application bioassay. All bioassays were conducted in the laboratory under

26

the same conditions as described for the rearing of insects, and commenced 2 h after

the onset of photophase. Insects were considered dead if they did not move any of

their appendages even when picked up with forceps. Mortalities were recorded every

24 h for 3-6 d depending on the bioassay. General descriptions of the bioassays are as

follows:

Petri dish exposure bioassay

Glass Petri dishes of the following dimensions were used in this bioassay: bottom

internal diameter 5 cm, sides height 1.3 cm and lid internal diameter 5.7 cm. The

surface area of the bottom, sides and lid is 19.6, 20.4 and 25.5 cm2, respectively. Test

samples in 1 ml of the appropriate solvent were prepared by dilution from stock

solutions. By using a 100 p\ syringe (Hamilton, Switzerland) 299, 312, and 389 p\

were applied to the bottom, sides and lid, respectively in order that all the treated

surface area contained the same sample concentration (wg/cm2). After sample

application to the bottom and the lid, they were distributed manually under a fume

hood during evaporation so that the test sample was deposited as uniformly as

possible over the bottom and the lid surfaces. The bottom was then held horizontally

for the application of the test sample to the sides and, after application, it was

continuously rolled, manually, over a flat surface under the fume hood until all the

solvent had evaporated, so that the test sample was deposited evenly over the sides.

The bottom and the lid were then left under the fume hood for lh in order to ensure

complete evaporation of solvent traces.

Groups of 10 randomly selected adult flies, 2-3 d-old, were immobilised by chilling

for a period of 5 min and subsequently introduced into each Petri dish, which was

27

then covered. Four thin cardboard strips (1 cm wide x 1.4 cm high, and less than

1 mm thick), oriented in four different directions, were placed between the sides and

the lid of each Petri dish so that the dishes were not airtight (Figure 2.4). Control

dishes were treated in a similar manner with solvent only, the same numbers of insects

were introduced into them and run simultaneously with the treatments. After 12 h

exposure to treated and control Petri dishes, flies of the same treatment were

transferred to wood-framed cages (12 x 12 x 12 cm) containing solid adult diet and

water (Figure 2.5). Prior to transfer, the Petri dishes were chilled briefly in order to

decrease the flies’ mobility, and consequently prevent insects from escaping them

during transfer. After each test the Petri dishes were thoroughly cleaned and placed in

the oven at 100 °C for 24 h.

Figure 2.4 A Petri dish used in the Petri dish exposure bioassay in which the test

sample was applied to the entire inner surface and thin cardboard strips were placed

between the sides and the lid.

28

Figure 2.5 Wood-framed cages, containing solid adult diet and water, in which tested

flies were transferred (10 flies/cage) after 12 h exposure to treated and control Petri

dishes.

The method described above involves two modifications compared to the standard

Petri dish exposure bioassay described by Abbassy et al. (1979): (a) the samples are

applied uniformly to all of the inner surface of the Petri dishes, instead of only to the

bottom, and (b) cardboard strips are placed between the sides and the lid of the Petri

dishes.

There are certain advantages to be gained using these modifications:

■ When a sample that possesses repellent properties is applied only to the

bottom of the Petri dish, the test insects have the potential to discriminate

between treated and untreated surfaces. Moreover, when a sample is applied

only to the bottom, there is a possibility that an external stimulus (e.g. light)

can influence the distribution of insects in the Petri dishes in a way such that

the insects prefer the untreated to the treated surfaces. In both cases, insects

acquire less amount of test sample and consequently their mortality will be

29

lower. In contrast, when the entire inner surface is lined with the test samples,

insects cannot avoid acquiring residues of the test samples from the Petri

dishes.

■ The application of the same amount of test samples gives 3.3 times the

concentration (wg/cm2) when applied only to the bottom than when applied to

the total inner surface. Consequently, in the case of crude extracts, in which

activity is not pronounced and/or the texture is sticky, the possibility of the

insects adhering to the extract film is reduced.

■ By using the cardboard strips between the sides and the lid the possibility that

mortality is due to fumigant action rather than contact is reduced, as the Petri

dishes were not air-tight and consequently volatiles could escape.

Note that the time required for the preparation of the Petri dishes using the two

modifications was much greater than that required for the unmodified apparatus.

Fumigation bioassay

This bioassay was designed in order to investigate the possibility that the observed

mortality in the Petri dish exposure bioassay was due to fumigation rather than

contact. For that purpose a sheet of gauze was fitted inside the Petri dishes to prevent

direct contact between deposit and insect. Moreover, in order to determine whether

the test samples have fumigant activity, a Petri dish fitted with a sheet of gauze was

sealed with Parafilm.

Four Petri dishes were prepared for each concentration (mg/Petri dish) tested. In one

Petri dish the test solution was applied to the total inner surface of the Petri dish as in

30

the Petri dish exposure bioassay, and in the other three Petri dishes only to the bottom.

A sheet of gauze was fitted inside two Petri dishes in which the test sample was

applied only to the bottom. Groups of 10 randomly selected adult flies, 2-3 d-old,

were immobilised by chilling and subsequently introduced into each Petri dish, which

was then covered. Cardboard strips were placed then between the sides and the lid of

the Petri dishes as in the Petri dish exposure bioassay except one in which the test

sample was applied to the bottom and a sheet of gauze was fitted. This Petri dish was

sealed with Parafilm. In Figure 2.6 the 4 Petri dishes described above are shown.

Control dishes were treated in a similar manner with solvent only, consisted of the

same numbers of insects and run simultaneously with the treatments. Tested flies were

Figure 2.6 The 4 Petri dishes used in the fumigation bioassay.

(A) Test sample was applied to the total inner surface and thin cardboard strips were placed between the sides and the lid; (B) Test sample was applied only to the bottom and thin cardboard strips were placed between the sides and the lid; (C) Test sample was applied only to the bottom, a sheet o f gauze was fitted inside the Petri dish and thin cardboard strips were placed between the sides and the lid; (D) Test sample was applied only to the bottom, a sheet o f gauze was fitted inside the Petri dish and sealed with Parafilm.

31

transferred in individual wood-framed cages (12 x 12 x 12 cm) containing solid adult

diet and water, 12 h after their introduction to the Petri dishes.

Topical application bioassay

In this study topical application bioassays were conducted according to the standard

method ‘number 1* described by McDonald et al. (1970). Groups of 10 randomly

selected adult flies, 2-3 d-old, were immobilised by chilling and subsequently

transferred into a Petri dish lined with filter paper. The Petri dish was then placed on

ice to prevent flies from reviving before treatment. One p\ of the test solution was

applied to the dorsal surface of the thorax of each fly with a hand micro-applicator

(Burkard, England) (Arnold, 1967) (Figure 2.7). The control run simultaneously

consisted of the same numbers of insects treated with 1 p\ of solvent each. After

Figure 2.7 The Burkard hand micro-applicator, used to apply topically 1 p\ of solvent

or test solution to the dorsal surface of the thorax of each fly.

32

treatment, flies were transferred in individual wood-framed cages (12 x 12 x 12 cm)

(10 flies/cage) containing solid adult diet and water.

2.2.3 Statistics

The program SPSS 8.0 for Windows provided the analysis environment for running

statistical tests (ANOVA, Tukey’s honestly significant difference test, probit analysis,

Pearson’s x 2 goodness-of-fit test, etc.) and transformations (arcsine, logio, etc.).

Comparison o f means

Mortality data were analysed by one-way analysis of variance (ANOVA) with

dose/concentration as the independent variable. Mortality was transformed to

arcsineV/? before ANOVA, where p is the proportion of dead insects, in order to

equalise the variance among treatments. Comparisons and separations of means were

performed by Tukey’s honestly significant difference (Tukey’s HSD) test (Tukey,

1953) aXP = 0.05.

Estimation o f median lethal dose concentration (LD5 &LC50)

In dose/concentration-response experiments the median lethal dose/concentration

(LD50/LC5 0) were estimated by probit analysis (Finney, 1971). In the case of control

mortality the recorded mortality (%) was corrected by Abbott’s formula (1925):

A4c = 100 x (Mp - Mt)/( 100 - Mt), where Me is the corrected mortality (%), Mp is the

recorded mortality (%) and Mt is the control mortality (%).

Dose can be defined as the amount of insecticide per unit weight or per insect. When

the exact dose given to the insect cannot be determined, then concentration is used

33

which can be defined as the amount of insecticide per unit of an external media

(Matsumura, 1975).

The median lethal dose/concentration (LD50/LC50) was in each case estimated from

the regression of the cumulative corrected probit mortality versus logio of the

doses/concentrations. The probit regression line allowed the estimation of LD50/LC50

and their corresponding 95 % confidence limits, and the slope and intercept with their

corresponding standard errors. Moreover, Pearson’s y 2 goodness-of-fit test was

performed in order to test how well the data fitted the probit model. When P > 0.05,

data fitted the probit model adequately. In contrast, when P < 0.05 data did not fit

adequately the probit model and a heterogeneity factor was automatically used in the

calculations of confidence limits by SPSS. LD5 0S/ LC50S were considered significantly

different when confidence limits did not overlap. From a biological point of view, the

slope of a probit regression estimates the change in activity per unit change in dose or

concentration (Robertson and Preisler, 1992). The length of 95 % confidence limits

provides a measurement of the precision of an LD /LC estimate. Shorter lengths

indicate greater precision (Finney, 1971) which allow differences between treatments

to be revealed. Precisely estimated LD50/LC50 are obtained when responses are evenly

distributed between 25 and 75 % (Robertson et al.y 1984).

34

CHAPTER 3

SCREENING FOR INSECTICIDAL ACTIVITY IN CITRUS

A URANTIUM EXTRACTS

3.1 INTRODUCTION

Variation in collection-site altitude, plant age, climate, and soil type can all influence

the concentration levels of secondary metabolites and even the kinds of compounds

biosynthesised in certain cases (Silva et al., 1998). Furthermore, the time of

collection, the time and temperature of extraction and the use of additional substances

during extractions are factors that can affect the chemical composition of plant

extracts (Houghton and Raman, 1998). Variation in the chemical profile of plant

material will probably cause a difference in the activity of extracts when tested

biologically (Houghton and Raman, 1998).

In the early stages of phytochemical investigations, one of the first and most obvious

questions to ask is whether the bioactive chemicals are localised to certain parts of the

plant (Cannell, 1998). Different organs in the plant are known to produce and/or

accumulate different profiles of secondary metabolites (Silva et al., 1998) and

probably their extracts can exhibit different degree of activity when tested against

insects. Gallo et al. (1996) reported that the crude hexane extract of Mammea

americana (Guttiferae) seeds was significantly more active against larvae of

Trichoplusia ni (Lepidoptera: Noctuidae) than the crude hexane extract of leaves

when an ingestion bioassay was used. Screening experiments conducted by Kadir et

35

al. (1989) with Spilanthes acmella (Compositae) leaf and flower head extracts against

adults of Periplaneta americana, using the topical application bioassay, revealed that

the active chemicals occurred in greater concentrations in the flower heads than in the

leaves. Weaver et al. (1994) examined the insecticidal activity of floral, foliar, and

root extracts of Tagetes minuta (Asterales: Asteraceae) against adult Mexican bean

weevils (Coleoptera: Bruchidae) using the Petri dish exposure bioassay. The results of

this study revealed that the root extract was the most active, followed by the flower

extract while the foliar extract was the least active.

Nogata et al. (1996) studied the inhibitory activity of different fruit tissue extracts

(flavedo, albedo, pulp and juice) of unripe and ripe fruits of Citrus lumia and Citrus

reticulata against cyclooxygenase and lipoxygenase, respectively. They reported that

the flavedo and the albedo extracts of Citrus lumia inhibited cyclooxygenase to the

same degree and were more inhibitory than the pulp extract, whereas the juice extract

had no effect on the enzyme. The flavedo, albedo and juice extracts of Citrus

reticulata unripe fruits inhibited lipoxygenase to the same degree and were more

inhibitory than the pulp. In contrast, the flavedo and the albedo extracts of Citrus

reticulata ripe fruits inhibited lipoxygenase to the same degree and were more

inhibitory than the juice extract, whereas the pulp extract exhibited slightly activity.

Variation in the chemical composition of plant samples can also occur due to

treatment after collection (Houghton and Raman, 1998). In most cases, plant material

is dried prior to extraction (Silva et al., 1998). Drying is the most common method of

preservation commercially and is achieved by leaving the material in warm, dry air

(Houghton and Raman, 1998). Temperature must not exceed 30 °C, and plant samples

36

must kept away from direct sunlight because ultraviolet radiation may produce

chemical reactions giving rise to compound artefacts (Silva et al., 1998). Oils

obtained from dried Citrus peels were reported to be much less toxic to the house fly

and red imported fire ant (Sheppard, 1984). However, Don-Pedro (1985)

demonstrated that powdered sun-dried orange and grapefruit peels applied as food

admixtures, exhibited toxic and repellent properties against Dermestes maculatus and

Callosobruchus maculatus.

Gbewonyo (1991) pointed out that it is important in studies involving the examination

of plants for insecticidal constituents, to test extracts obtained with different solvents

and methods. Ideally, an extraction solvent should be easy to remove, and be inert,

non-toxic, and not easily flammable (Silva et al., 1998). The nature of the extraction

solvent determines the type o f chemicals it is likely to extract based on the “like

dissolves like” general principle; non-polar solvents will extract out non-polar

substances, and polar materials will be extracted out by polar solvents (Houghton and

Raman, 1998). Salvatore et al. (2004) compared the activity of lemon peel extracts

obtained with different solvents using an ultrasonic bath (20 °C for 20 min) by a larval

toxicity bioassay against CC. They found that the diethyl ether extract was the most

toxic, followed by the methanol extract, whereas the ethyl acetate extract was the least

toxic.

Morgan and Wilson (1985) emphasised that extraction methods involving heating are

inappropriate for the isolation of some insecticidal components from plants since

some of these compounds may be heat-labile. This view was supported by Gbewonyo

(1991) who reported that although Soxhlet extraction of the male root of Piper

37

guineense gave more extraction yield, its activity against Musca domestica by the

topical application method, was less than the extract obtained by cold extraction.

The fact that plant secondary metabolite profiles may vary both quantitatively and

qualitatively among different batches of the same plant part collected at different

times has important consequences in making plant recollections for scale-up work

(Silva et al., 1998). Chloroform-methanol extracts of ripe and unripe fruits of Piper

guineense did not differ significant in activity by the topical application bioassay

against Musca domestica and Periplaneta americana adults although the activity of

the ripe fruit extract was greater against both insects (Gbewonyo, 1991). The

inhibitory activity against cyclooxygenase of Citrus lumia pulp extract decreased

during ripening, while there was little difference in extract inhibition between unripe

and ripe flavedo, albedo and juice extracts (Nogata et al., 1996). Moreover, the

inhibitory activities of Citrus reticulata pulp and juice extracts declined during

ripening, while the activities of flavedo and albedo extracts remained almost constant.

Even though it is well documented that Citrus species contain chemicals that exhibit a

wide range of insecticidal properties against a broad spectrum of insect pests, CA

among other Citrus species had received little attention. That fact suggests that the

investigation of the insecticidal actions of CA is worthwhile.

The objectives of this component of the study were:

■ To screen various solvent fractions obtained from different plant parts of CA

for insecticidal activity against adults of BO.

38

■ To determine whether the active chemicals present in the fruits are localised to

specific fruit tissues.

■ To study the effect of different drying treatments on the stability of the

bioactive chemicals present in the peels.

■ To compare the effectiveness of different extraction methods and solvents on

the recovery of the insecticidal chemicals.

■ To study the effect of fruit maturation on the biosynthesis and/or

bioaccumulation of the active chemicals.

39

3.2 METHODS

3.2.1 Extraction of plant materials

Plant materials from the selected CA trees were collected at different dates. In each

collection date the same number of fruits, leaves and shoots were collected from each

of the 10 sampling trees. The samples were transferred immediately to the laboratory,

cleaned and extracted fresh unless otherwise stated. The collection data for CA

samples are shown in Table 3.1.

Table 3.1 Collection data for CA samples.

Collectiondate Plant part Plant part

weight (g)Plant part description

Number of plant parts collected

01/08/1998 Fruits 30-40 Dark green 2001/08/1998 Leaves 1.0-1.5 Annual 160

01/08/1998 Shoots 4-6 Annual 40

20/08/1998 Fruits 40-60 Dark green 50


01/10/1998 Fruits 60-100 Light green 60



01/10/1999 Fruits 60-100 Light green 30

01/11/1999 Fruits 70-120 Green-yellow 30

01/12/1999 Fruits 80-130 Yellow-orange 30

01/01/2000 Fruits 90-140 Orange 30

01/02/2000 Fruits 90-140 Orange 30

Extraction o f CA plant parts

Fruits, leaves and shoots of CA (Figure 3.1) were collected on 01/08/98 in order to

investigate whether these plant parts contain chemicals with insecticidal activity. Ten

40

Shoots

Leaves

Fruits

Figure 3.1 Fruits, leaves and shoots of CA.

41

fruits ( 1 from each tree), 80 leaves ( 8 from each tree) and 2 0 shoots ( 2 from each tree)

weighing 350, 100 and 100 g, respectively were cut into small pieces and

homogenised separately with methanol in the blender. A portion of the homogenised

fruits equivalent to 1 0 0 g fruits and the homogenised leaves and shoots were extracted

separately by the conventional solid-liquid extraction method using methanol as the

extraction solvent.

Methanol was used as the initial extraction solvent as it can extract compounds

representing a wide polarity range and is a so-called ‘all-purpose’ solvent (Cannell,

1998). Furthermore, methanol efficiently penetrates cell membranes, permitting the

extraction of high amounts of endocellular components and also prevents, by

denaturing the plant enzymes responsible, enzymatic processes or reactions that start

after the plant material is collected (Silva et al., 1998).

The resulting mixtures were filtered through a double-layer Gen-purpose filter paper

(Schleicher & Schuell, Germany) on a Buchner funnel with a slight suction and the

methanol extracts (filtrates) were concentrated to dryness under reduced pressure at

35 °C in a rotary evaporator (a Rotavapor R-l 14 fitted with a waterbath B-480, Buchi,

Switzerland). The partitioning procedure of CA methanol extracts is displayed in

Figure 3.2. The dried methanol extracts were suspended in 100 ml water and

partitioned successively using separatory funnels, 3 times with 100 ml of petroleum

ether (b.p. 40-60 °C), dichloromethane and ethyl acetate. After gravity filtration and

solvent evaporation, the petroleum ether (MeOH-PE), dichloromethane (MeOH-

DCM) and ethyl acetate (MeOH-EtOAc) fractions of the fruit, leaf and shoot

methanol extracts were obtained. The remaining aqueous layers were concentrated in

42

Homogenised plant material of CA (100 g fresh)

Extracted with 1 1 of MeOH for 24 h Filtered

Methanol extractConcentrated

in vacuum Dried methanol extract

Suspended in 100 ml H20 Liquid-liquid partitioning (100 ml H20 / 3x100 ml PE)

PE layerFiltered Concentrated

in vacuum MeOH-PE fraction

Aqueous layerLiquid-liquid partitioning (100 ml H20 / 3x100 ml DCM)

DCM layerFiltered Concentrated

in vacuumMeOH-DCM fraction

Aqueous layerLiquid-liquid partitioning (100 ml H20 / 3x100 ml EtOAc)

EtOAc layer Filtered Concentrated

in vacuumMeOH-EtOAc fraction

Aqueous layerConcentrated

in vacuum Resuspended in MeOH (3x100 ml) Stirred Filtered Concentrated

in vacuumMeOH-Residual fraction

Figure 3.2 Partitioning procedure of CA methanol extracts.

the rotary evaporator, resuspended in methanol (3x100 ml), stirred each time for 5

min in the Btichi rotavapor without applying suction, gravity filtered and finally

concentrated, yielding the residual fraction (MeOH-Residual) of the fruit, leaf and

shoot methanol extracts.

The above extraction strategy was used in preference to a series of stepwise

extractions (exhausted extraction o f plant material by using successively extraction

solvents of increasing polarity) as it is less time- and solvent-consuming.

The 10 remaining fruits were cut in half and, together with the remaining leaves (n =

80) and shoots (n = 2 0 ) which were cut into small pieces, were dried to constant

weight under darkness at 50 °C in a well-ventilated oven (B50, Memmert, W.

Germany) and their percentage dry weight was estimated. Percentage dry weight was

used in order to express the extraction yields as milligram extractable material per

gram of dried plant material extracted.

Extraction o f CA fruit tissues

Fruits of CA were collected on 20/08/98 in order to determine whether the insecticidal

active chemicals present in the MeOH-PE fraction of the fruits are localised to a

specific fruit tissue. Leaves and shoots of CA were not further examined as all their

fractions were inactive.

Twenty fruits (2 from each tree) were dissected into peels (flavedo), albedo and flesh

(Figure 3.3). The peels, the albedo, the flesh and 10 intact (whole) fruits (1 from each

tree) weighing 100, 100, 320 and 500 g, respectively were cut into small pieces and

44

Figure 3.3 Fruit tissues of CA.

homogenised separately with methanol. A portion of the homogenised fruits and flesh

equivalent to 1 0 0 g fruits and flesh, respectively and the homogenised peels and

albedo were extracted separately by the conventional solid-liquid extraction method

using methanol as the extraction solvent. The resulting mixtures were suction-filtered

and then the methanol extracts (filtrates) were concentrated to dryness under reduced

pressure at 35 °C in the rotary evaporator. The dried methanol extracts were then

suspended in 100 ml water and partitioned only with petroleum ether (3 times with

45

1 0 0 ml each time), yielding after gravity filtration and solvent evaporation the

petroleum ether fractions of the fruit, peel, albedo and flesh methanol extracts. The

remaining aqueous layers after their partitioning with petroleum ether were discarded

as the dichloromethane, ethyl acetate and residual fractions of the fruit methanol

extract were inactive.

Ten fruits (1 from each tree) were cut in half and placed in the oven at 50 °C in order

to estimate their percentage dry weight. The 10 remaining fruits were dissected into

peels, albedo and flesh in order to estimate their percentage dry weight.

Extraction o f fresh and dried peels o f CA

Fruits of CA were collected on 10/09/98 in order to investigate whether drying of

peels prior to extraction can affect the insecticidal activity of their MeOH-PE

fractions. Intact fruits, albedo and flesh were not further studied as the activity of the

MeOH-PE peel fraction was significant greater than that of the whole fruit, whereas

the albedo and flesh fractions were inactive.

The bulk sample of peels obtained by knife-peeling of the fruits was divided into 3

sub-samples (100 g each). One sub-sample was randomly selected to be extracted

fresh and the other two sub-samples were randomly assigned to the two oven-drying

treatments (30 °C for 4 d and 50 °C for 2 d). Oven-drying was preferred than sun

drying as ultraviolet radiation may produce chemical reactions giving rise to

compound artefacts (Silva et al., 1998).

46

The fresh and the two dried peel sub-samples were extracted separately using the

conventional solid-liquid extraction method and methanol as the extraction solvent.

The petroleum ether fractions of the fresh and the two dried peel methanol extracts

were obtained as in the extraction of the fruit tissues.

Extraction o f CA peels by different extraction methods and solvents

Fruits of CA were collected on 01/10/98 in order to investigate the effect o f two

different extraction methods and three different initial extraction solvents on the

insecticidal activity of peel fractions and extracts obtained by the extraction of fresh

peels. Dried peels were not further examined, as their MeOH-PE fractions were

inactive.

The ultrasound and the conventional solid-liquid extraction methods were compared.

The ultrasound extraction method was selected in order to be compared with the

conventional solid-liquid extraction method, as it is more rapid and it employs less

solvent than the conventional solid-liquid extraction (Poole and Poole, 1991).

Furthermore, the efficacy of methanol, water and petroleum ether, as initial extraction

solvents was compared with both extraction methods. Water was used in order to

investigate whether it can extract the non-polar insecticidal active compounds present

in the MeOH-PE fraction. Petroleum ether was used in order to investigate whether

the concentration of the insecticidal active compounds was greater in the petroleum

ether extracts than in the petroleum ether fractions of the methanol extracts.

The bulk sample of peels obtained by knife-peeling of the fruits was divided into six

sub-samples (100 g each). Three sub-samples were randomly assigned to be extracted

47

separately with petroleum ether, methanol and water as initial extraction solvents

using the conventional solid-liquid extraction method. The other 3 sub-samples were

randomly assigned to be extracted with the three different initial extraction solvents

using the ultrasound extraction method. The methanol extracts obtained from the two

extraction methods were partitioned with petroleum ether as in the extraction of fruit

tissues, yielding the conventional solid-liquid petroleum ether (Conv. S-L MeOH-PE)

and the ultrasound petroleum ether (Ultrasound MeOH-PE) fractions of the methanol

extracts. The aqueous extracts from each extraction method were concentrated to 100

ml and then partitioned with petroleum ether (3x100 ml), yielding the conventional

solid-liquid petroleum ether (Conv. S-L H2O-PE) and ultrasound petroleum ether

(Ultrasound H2O-PE) fractions of the aqueous extracts. The petroleum ether extracts

from both extraction methods were suction-filtered and concentrated to dryness,

yielding the conventional solid-liquid petroleum ether (Conv. S-L PE) and ultrasound

petroleum ether (Ultrasound PE) extracts.

Extraction o f peels obtainedfrom CA fruits at different ripening (maturity) stages

Seven collections of fruits of CA with a month interval starting on 01/08/99 (3 months

after fruit set) until full maturity (last collection on 0 1 /0 2 /0 0 ) were performed in order

to investigate whether there is within-seasonal variation in the insecticidal activity of

the petroleum ether peel extracts.

The bulk sample of peels of each collection was divided in two sub-samples (100 g

each) which were randomly assigned to be extracted and dried. The peels were

extracted by using the ultrasound extraction method and petroleum ether as the

extraction solvent, yielding the petroleum ether peel extracts.

48

Although the insecticidal activity of the peel extracts obtained by both methods using

petroleum ether as the extraction solvent was similar, peels were extracted by the

ultrasound method as this method is less time- and solvent-consuming than the

conventional solid-liquid extraction method. Methanol and water were not further

used as initial extraction solvents since the H2O-PE fractions of both methods and the

ultrasound MeOH-PE fraction were inactive, whereas the petroleum ether extracts of

both methods exhibited significantly greater insecticidal activity than the conventional

solid-liquid MeOH-PE fraction.

The weight (mg) of all the fractions and extracts was estimated by concentrating to

near-dryness, an appropriate portion or the whole fraction or extract in the rotary

evaporator at 35 °C. The portion o f each fraction or extract was selected in order to

weigh 50-100 mg, on the basis of preliminary weighing of 5 % of the whole fraction

or extract. The concentrated fraction or extract was subsequently transferred in a 15-

ml vial (15-ml clear glass vial, screw top solid cap with PTFE liner, Supelco, USA)

immersed in a water bath at 35 °C and concentrated to constant weight under a stream

of nitrogen. All weighing was performed on an electronic four-figure balance (AB54-

S, Mettler Toledo, Switzerland). All fractions and extracts were concentrated to

dryness and stored in the dark at -20 °C until used.

3.2.2 Bioassay of fractions and extracts

The Petri dish exposure bioassay was used in order to evaluate the insecticidal activity

of different fractions and extracts of CA against 2-3 d-old adults of BO. Five

replicates of 10 BO flies each (5 males and 5 females) were employed for each

fraction/extract concentration and the control. The 5 replicates were conducted on 5

49

successive days, utilising the same generation (batch) of flies. Different

concentrations of screened fractions/extracts were not run simultaneously as screening

experiments at lower concentrations included only the more active fractions/extracts.

Stock solutions were prepared in DCM, as were subsequent dilutions except of the

plant parts fractions, which were prepared in a mixture of MeOH:DCM (3:1, v/v).

Fresh test solutions were prepared for each replicate. The syringe used for the

application of samples was thoroughly cleaned after the application of each sample.

Mortalities were recorded every 24 h for 4 d after which there was no further increase

in mortality.

3.2.3 TLC analysis of fractions and extracts

Thin layer chromatography (TLC) is an extremely sensitive analytical method which

can analyze a large number of samples simultaneously, although it does not require

special and expensive equipment (Touchstone and Dobbins, 1983). TLC analyses

were performed on 20 x 20 cm Sigma glass plates pre-coated with 250 //m silica gel

60 F2 5 4 . CA fractions and extracts were spotted (100 //g in 5 fi\ of the appropriate

solvent) using a 10 (A Hamilton syringe onto the origin line (a line marked 2 cm from

one end of the plate). All fractions and extracts were dissolved in DCM, except the

MeOH-EtOAc, and MeOH-Residual fractions of fruits, leaves and shoots which were

dissolved in MeOH. A glass chamber (29 cm long x 10 cm wide x 27 cm high,

Sigma-Aldrich, USA) was used for plates’ development. Prior to the introduction of

the appropriate developing solvent system, the inner surface of the chamber was lined

with filter paper to aid in saturating the atmosphere with the solvent vapours. The

appropriate developing solvent system was used in each case in order to give the best

possible separation based on preliminary analyses. After introducing the developing

50

solvent system (100 ml) to the glass chamber an interval of 30 min was allowed for

saturation of the atmosphere with the solvent vapours. TLC plates were developed in

the saturated chamber to a distance of 15 cm. After development, the TLC plates were

air-dried for 1 h at room temperature under a fume hood. Photographs of the TLC

plates were obtained by a Camedia digital camera (C-2000Z, Olympus, Japan) under

UV light of 254 nm provided by a Spectroline lamp (ENF-280C/F, Spectronics

Corporation Westbury, USA). Under UV light of 254 nm eluted components present

in CA fractions and extracts were visible as dark spots or areas against a greenish

fluorescent background.

51

3.3 RESULTS

3.3.1 Extraction yield, chemical profile and insecticidal activity of various

fractions of CA plant parts

Petroleum ether, dichloromethane, ethyl acetate and residual fractions of the methanol

extracts of CA fruits, leaves and shoots were obtained as described in section 3.2.1 in

order to investigate whether these fractions contain insecticidal active chemicals.

Extraction yield

The extraction yield of all fractions and their contribution to the sum of fractions of

the same plant part are shown in Table 3.2. The residual fractions accounted for 79.0-

87.8 % o f the sum of fractions of the same plant part, whereas the petroleum ether,

dichloromethane and ethyl acetate fractions attained only 1.8-17.0,1.6-4.7 and 2.0-6.3

%, respectively. The chemicals present in the petroleum ether fractions exhibited a

wide range of contribution to the sum of fractions of the same plant part, represented

1.8, 9.2 and 17.0 % of the sum of fruit, shoot and leaf fractions, respectively.

Table 3.2 Extraction yield of various fractions of CA plant parts and % contribution

of each fraction to the sum of fractions of the same plant part.

Fraction8Plant part

Fruit Leaf Shoot Fruit Leaf ShootExtraction yield (mg/g ffesh)b Contribution (%)

MeOH-PE 1.80 16.65 4.92 1.8 17.0 9.2MeOH-DCM 4.72 1.97 0.87 4.7 2.0 1.6

MeOH-EtOAc 5.78 1.91 3.33 5.7 2.0 6.3

MeOH-Residual 88.32 77.28 44.12 87.8 79.0 82.9

*MeOH-PE, MeOH-DCM, MeOH-EtOAc and MeOH-Residual: Petroleum ether, dichloromethane, ethyl acetate and residual fractions, respectively. ‘̂ Extraction yield of fractions was recorded as fractions weight (mg) per g of fresh plant material extracted.

52

The high contribution of the residual fractions to the sum of fractions of the same

plant part was expected as the initial extraction solvent used was methanol. Methanol

extracts mainly polar compounds, based on the “like dissolve like” general principle.

These polar compounds cannot be extracted by the solvents used (petroleum ether,

dichloromethane and ethyl acetate) during the liquid-liquid partitioning of CA

methanol extracts and thus remained in the residual fractions.

The dry weight (%) of CA fruits, leaves and shoots and the extraction yield of the sum

of fractions of each plant part are shown in Table 3.3. The dry weight (%) of shoots

was 1.2 and 2.3 times greater than that of the leaves and fruits, respectively. The

extraction yield of the sum of fractions of fruits and leaves when expressed as

milligrams per gram of fresh plant material extracted was similar and at least 1 . 8

times greater than that o f the shoots. However, when extraction yield was expressed

as milligrams per gram of dried plant material extracted, the extraction yield of fruits

was 2-fold and 4.4-fold greater than that of the leaves and shoots, respectively. This

finding indicates that the differences among extraction yields of the different plant

Table 3.3 Dry weight (%) of CA plant parts and extraction yield of their sum of

fractions.

Plant part Fresh amount extracted (g)

Dry weight(%)

Sum of fractionsExtraction yield

mg/g fresh8 mg/g driedb

Fruit 100 24.7 100.62 407.37Leaf 100 48.6 97.81 201.26

Shoot 100 57.2 53.24 93.08

“Extraction yield of the sum of fractions was recorded as sum of fractions weight (mg) per g of fresh plant material extracted. '’Extraction yield of the sum of fractions was recorded as sum of fractions weight (mg) per g of dried plant material extracted.

53

parts were not due to the dry weight extracted (equivalent to 1 0 0 g fresh), but rather to

the nature of the secondary metabolites present in each plant part.

Chemical profile

The various fractions of CA fruits, leaves and shoots were analysed by TLC in order

to compare their chemical profiles. Different developing solvent systems were

employed for the analyses of these fractions as secondary metabolites present in them

represented a wide range of polarity. Preliminary analyses showed that the most

appropriate developing solvent system in order to analyse simultaneously the four

fractions of the three plant parts of CA was ethyl acetate (100 %). Comparison among

chemical profiles of dichloromethane fractions was performed using this developing

solvent system. By using Hex.EtOAc (50:50, v/v) as the developing solvent system

the petroleum ether fractions were compared. Furthermore, the ethyl acetate and the

residual fractions which contained mainly polar compounds were analysed by using

as developing solvent system DCM:MeOH (85:15, v/v) and DCM:MeOH (80:20,

v/v), respectively. The solutions used to spot the TLC plates were prepared in DCM

for the petroleum ether and dichloromethane fractions, while the ethyl acetate and

residual fraction were prepared in MeOH.

In Figure 3.4 the analysed fractions (the four fractions of the three plant parts of CA)

using EtOAc (100 %) as the developing solvent system are shown. Great quantitative

and qualitative differences were found among petroleum ether, dichloromethane, ethyl

acetate and residual fractions of the same plant part. In Figures 3.5A, 3.5B and 3.5C

the analysed petroleum ether, ethyl acetate and residual fractions of fruits, leaves and

shoots of CA using Hex:EtOAc (50:50, v/v), DCM:MeOH (85:15, v/v) and

54

DCM:MeOH (80:20, v/v) as developing solvent systems, respectively are shown.

Quantitative and quantitative differences were found among dichloromethane

fractions (Figure 3.4), petroleum ether fractions (Figure 3.5A), ethyl acetate fractions

(Figure 3.5B) and residual fractions (Figure 3.5C), of fruits, leaves and shoots of CA.

*r1.00 —

0.95 -

0.90 —

0.85 -

0.80 —

0 .7 5 -

0.70 —

0.65 -

0.60 —

0.55 -

0.50 —

0.45 -

0.40 —

035 -

0.30 —

0.25 -

0.20 —

0 . 1 5 -

0.10 —

0.05 -

0.00 —

FI LI SI F2 L2 S2 F3 L3 S3 F4 L4 S4

Figure 3.4 Thin layer chromatogram of various fractions of CA plant parts.

FI, F2, F3 and F4: Petroleum ether, dichloromethane, ethyl acetate and residual fruit fractions, respectively; LI, L2, L3 and L4: Petroleum ether, dichloromethane, ethyl acetate and residual leaf fractions, respectively; SI, S2, S3 and S4: Petroleum ether, dichloromethane, ethyl acetate and residual shoot fractions, respectively; R f: Retention factor. Developing solvent system: 100 % EtOAc. UV light: 254 nm. Amount o f fractions per spot: 100 fig.

The differences in the range and proportions of chemical substances among the

different fractions of the same plant part indicate that the fractionation procedure

(liquid-liquid partitioning with solvents of increasing polarity) of the methanol

extracts was effective, as different types of compounds based on their polarity were

55

present in each different fraction. The quantitative and qualitative differences among

fruits, leaves and shoots of the same solvent fraction indicate that fruits, leaves and

shoots of CA produce and/or accumulate different profiles of secondary metabolites.

F L S F L S F L S

Figure 3.5 Thin layer chromatograms of petroleum ether (A), ethyl acetate (B) and

residual (C) fractions of CA plant parts.

F: Fruit; L: Leaf; S: Shoot; R f: Retention factor. Developing solvent systems: (A) Hex:EtOAc (50:50, v/v); (B) DCM:MeOH (85:15, v/v); (C) DCM:MeOH (80:20, v/v). UV light: 254 nm. Amount o f fractions per spot: 100

Insecticidal activity

The various fractions of CA fruits, leaves and shoots were screened for insecticidal

activity against BO adults by the Petri dish exposure bioassay. As these fractions

contained chemicals with a wide range of polarity a solvent mixture of MeOH:DCM

56

(3:1, v/v), which possesses the solubility characteristics to dissolve all fractions was

employed. By finding one solvent mixture which can dissolve all fractions, only one

control was used which was treated with this solvent mixture only. The screening

concentration was set at 200 //g/cm2 and so 13.1 mg (200 pg/cm2 x 65.5 cm2) of each

fraction in 1 ml (0.75 ml MeOH and 0.25 ml DCM) was applied to the total inner

surface of the Petri dishes as described in section 2.2.2.

The results of this experiment are displayed in Table 3.4. The petroleum ether fraction

Table 3.4 Percentage mortality of BO adults exposed to various fractions of CA plant

parts at 200 //g/cm2 using the Petri dish exposure bioassay.

Fraction8% Mortality

Meanb ± SEC, at hours after treatment

24 48 72 96

Fruit MeOH-PE 16 ± 4.0a 60 ± 5.5a 74 ± 5.1a 76 ± 5.1aFruit MeOH-DCM 0 ± 0.0b 0 ± 0.0b 2 ± 2.0b 2 ± 2.0b

Fruit MeOH-EtOAc 0 ± 0.0b 2 ± 2.0b 2 ± 2.0b 2 ± 2.0b

Fruit MeOH-Residual 0 ± 0.0b 0 ± 0.0b 0 ± 0.0b 0 ± 0.0b

Leaf MeOH-PE 0 ± 0.0b 0 ± 0.0b 0 ± 0.0b 0 ± 0.0b

Leaf MeOH-DCM 0 ± 0.0b 0 ± 0.0b 0 ± 0.0b 0 ± 0.0b

Leaf MeOH-EtOAc 0 ± 0.0b 0 ± 0.0b 0 ± 0.0b 0 ± 0.0b

Leaf MeOH-Residual 2 ± 2.0b 4 ± 2.4b 4 ± 2.4b 4 ± 2.4b

Shoot MeOH-PE 0 ± 0.0b 0 ± 0.0b 0 ± 0.0b 0 ± 0.0b

Shoot MeOH-DCM 0 ± 0.0b 4 ± 2.4b 4 ± 2.4b 4 ± 2.4b

Shoot MeOH-EtOAc 0 ± 0.0b 0 ± 0.0b 0 ± 0.0b 0 ± 0.0b

Shoot MeOH-Residual 0 ± 0.0b 0 ± 0.0b 2 ± 2.0b 2 ± 2.0b

Control41 0 ± 0.0b 0 ± 0.0b 0 ± 0.0b 0 ± 0.0b

*MeOH-PE, MeOH-DCM, MeOH-EtOAc and MeOH-Residual: Petroleum ether, dichloromethane, ethyl acetate and residual fractions, respectively. bMean o f 5 replicates o f 10 insects each (5 males and 5 females). CSE: Standard error o f mean. dTreated with a solvent mixture o f MeOH:DCM (3:1, v/v). Means within a column followed by the same letter are not significantly different (P = 0.05, Tukey’s HSD test). Mortalities were transformed before ANOVA to arcsine Vp, where p is the proportion o f dead insects. Means ± SE o f untransformed data are reported.

57

of the fruit methanol extract was the only fraction which exhibited activity against BO

adults (76 % mortality 96 h after treatment). Its activity was significantly greater from

all other fractions and the control at all mortality recording times.

The results indicate that the MeOH-PE fraction of the fruits contained one or more

chemical substances that can kill BO adults. All other fractions either did not contain

insecticidally active chemicals or contained them in such proportions that are not

toxic at the screening concentration used.

3.3.2 Extraction yield, chemical profile and insecticidal activity of petroleum

ether fractions of CA fruit tissues

Petroleum ether fractions of the methanol extracts of CA intact fruits, peels, albedo

and flesh were obtained as described in section 3.2.1 in order to determine whether a

specific fruit tissue was mainly responsible for the insecticidal activity of the

petroleum ether fraction of the fruit methanol extract.

Extraction yield

The dry weight (%) of CA fruit tissues and the extraction yield of their petroleum

ether fractions are displayed in Table 3.5. The dry weight (%) of peels and albedo was

similar and at least 1.4 and 1.9 times greater than that of the intact fruits and flesh,

respectively. The extraction yield of the petroleum ether fractions of the peels and

flesh when expressed as milligrams per gram of fresh fruit tissue extracted was

approximately the same and at least 1.2 and 2.5 times respectively greater than that of

the intact fruits and the albedo. However, when the extraction yield was expressed as

milligrams per gram of dried fruit tissue extracted, the petroleum ether fraction of

58

intact fruit was 1.2 and 3.0 times greater than that of the peels and albedo,

respectively but 1.7 times less than that of the flesh.

Table 3.5 Dry weight (%) of CA fruit tissues and extraction yield of their petroleum

ether fractions.

Fruit tissue Fresh amount extracted (g)

Dry weight(%)

Petroleum ether fractionsExtraction yield

mg/g fresh8 mg/g driedb

Intact fruit 100 24.4 2.02 8.28

Peel 100 35.0 2.47 7.06

Albedo 100 35.8 0.99 2.77

Flesh 100 18.2 2.50 13.74

“Extraction yield o f fractions was recorded as fractions weight (mg) per g o f fresh fruit tissue extracted. ’’Extraction yield o f fractions was recorded as fractions weight (mg) per g o f dried fruit tissue extracted.

These findings indicate that the observed differences in the extraction yield among

petroleum ether fractions of CA fruit tissues were not due to their dry weight

extracted (equivalent to 1 0 0 g fresh), but rather to the chemical composition of these

fruit tissues.

Chemical profile

Comparison of the chemical profiles of the petroleum ether fractions of CA fruit

tissues was performed by TLC using Hex:EtOAc (50:50, v/v) as the developing

solvent system. By using this single developing solvent system the petroleum ether

fractions were analysed adequately as they contained chemicals of similar polarities

(mainly non-polar chemicals).

59

In Figure 3.6 the analysed petroleum ether fractions of CA fruit tissues using

Hex:EtOAc (50:50, v/v) as developing solvent system are shown. Great quantitative

and qualitative differences were revealed among the three fruit tissue fractions.

Among the four fractions analysed, the peel and the intact fruit chemical profile look

more similar than any other pair of fractions, although that between them quantitative

and qualitative differences were evident. TLC analysis indicates that different fruit

tissues of CA produce and/or accumulate different profiles of secondary metabolites.

*r1.00 —

0 .9 5 -

0.90 — 0 . 8 5 -

0.80 — 0 . 7 5 -

0.70 — 0 .6 5 -

0.60 — 0.55 -

0.50 — 0 .4 5 -

0.40 — 0 .3 5 -

0.30 — 0 . 2 5 -

0.20 —

0 . 1 5 -

0.10 —

0.05 -

0.00 —

Figure 3.6 Thin layer chromatogram of petroleum ether fractions of CA fruit tissues.

Fr, P, A and FI: Petroleum ether fractions o f intact fruit, peel, albedo and flesh, respectively; R t: Retention factor. Developing solvent system: Hex:EtOAc (50:50, v/v). UV light: 254 nm. Amount o f fractions per spot: 100 fxg.

60


The petroleum ether fractions of CA fruit tissues were tested against BO adults, using

the Petri dish exposure bioassay, in order to investigate in which fruit tissue fraction

the insecticidal active chemicals were present in greater concentrations. The screening

concentration used for the initial experiment was 200 //g/cm2. Only fractions that gave

significant activity at this concentration were further tested at 1 0 0 //g/cm2.

The results of these experiments are shown in Table 3.6. The albedo and the flesh

fractions were inactive ( 2 - 4 % mortalities 96 h after treatment) at 200 //g/cm2. At the

same concentration the mortality of the peel fraction was greater than that of the intact

fruit fraction at all mortality recording times. However, only at 72 and 96 h after

treatment were those differences significant. Furthermore, the peel fraction caused

Table 3.6 Percentage mortality o f BO adults exposed to petroleum ether fractions of

CA fruit tissues using the Petri dish exposure bioassay.

Concentration/ig/cm2

Fruittissue

% MortalityMean8 ± SEb, at hours after treatment

24 48 72 96

Intact fruit 26 ± 10.3a 62 ± 5.8a 68 ± 8.0b 70 ± 7.1bPeel 46 ± 11.7a 82 ± 8.6a 96 ± 2.4a 100 ± 0.0a

200 Albedo 0 ± 0.0b 0 ± 0.0b 0 ± 0.0c 2 ± 2.0c

Flesh 2 ± 2.0b 2 ± 2.0b 4 ± 2.4c 4 ± 2.4c

Control0 0 ± 0.0b 0 ± 0.0b 0 ± 0.0c 0 ± 0.0c

Intact fruit 0 ± 0.0b 12 ± 4.9b 14 ± 4.0b 14 ± 4.0b100 Peel 14 ± 7.5a 66 ± 12.1a 80 ± 10.5a 84 ± 6.8a

Control 0 ± 0.0b 0 ± 0.0b 0 ± 0.0c 0 ± 0.0c

“Mean o f 5 replicates o f 10 insects each (5 males and 5 females). bSE: Standard error o f mean. 'Treated with DCM. The two experiments (200 and 100 //g/cm2) were not run simultaneously. Means within a column o f a single concentration followed by the same letter are not significantly different (P = 0.05, Tukey’s HSD test). Mortalities were transformed before ANOVA to arcsine V/?, where p is the proportion o f dead insects. Means ± SE o f untransformed data are reported.

61

significantly more mortality to BO adults than the intact fruit fraction at all mortality

recording times at 1 0 0 //g/cm2.

The results indicate that the peel fraction possesses, in greater proportions, the

insecticidally active chemical compounds present in the intact fruit fraction. The

albedo and flesh fractions either did not contain these chemicals or contained them in

such proportions that cannot exhibit any evidence of activity even at 200 /^g/cm . It

seems that the bioactive chemicals present in the intact fruit fraction are mainly

produced and/or accumulated in the peel tissue of CA fruits.

3.3.3 Extraction yield, chemical profile and insecticidal activity of petroleum

ether fractions obtained from fresh and dried peels of CA

Petroleum ether fractions of the methanol extracts obtained by the extraction of fresh

and dried peels were prepared as described in section 3.2.1 in order to investigate

whether chemical and/or enzymatic processes or reactions possibly occurring during

the two drying processes of the peels can change the insecticidal activity of their

petroleum ether fraction.

Extraction yield

The extraction yield of the petroleum ether peel fractions obtained from fresh, oven-

dried at 30 °C for 4 d and oven-dried at 50 °C for 2 d peels, recorded as milligrams per

gram fresh peel was 2.66, 1.60 and 1.76, respectively. The extraction yield of the two

petroleum ether peel fractions obtained from dried peels was similar and at least 1.5

times less than that of the petroleum ether fraction obtained from fresh peels.

62

The reduced extraction yields of the two fractions obtained from dried peels could be

attributed to the transformation, during drying treatments, of some chemicals present

in the fraction obtained from fresh peels, to more polar chemicals due to enzymatic

and/or chemical processes or reactions. These artefacts possibly cannot be extracted

from the methanol extracts by the liquid-liquid partitioning with petroleum ether, due

to their greater polarity.

Chemical profile

Comparison of the chemical profiles of the petroleum ether peel fractions obtained

from fresh and oven dried peels was performed by TLC using Hex:EtOAc (50:50,

v/v) as the developing solvent system. Figure 3.7 shows the analysed petroleum ether

peel fractions of CA. TLC analysis revealed that there were great similarities among

the chemical profiles o f the three fractions. However, the two fractions obtained from

dried peels were almost the same (only quantitative difference between spots with R(.

0.30), and they differ with the fraction obtained from fresh peels only quantitatively

(spots with R{. 0.00, 0.12, 0.19, 0.30 and 0.46).

The quantitative differences revealed when the fraction obtained from fresh peels was

compared with the two fractions obtained from dried peels could be attributed to the

transformation of some chemicals or part of them to more polar chemicals during

drying treatments. However, the greater proportions of the chemicals with Rf. 0.12,

0.19 and 0.30 in the fractions obtained from dried peels compared to the fraction

obtained from fresh peels could be explained by the lower yield of these fractions

rather than the greater amount of these chemicals in the fractions. Furthermore, the

63

small difference between the fractions obtained from dried peels must have been due

to the effect of the different temperatures and exposure times of the drying treatments.

1.0 0 -

0.95

0 .90 -0.85

0 .80 -0.75

0 .70 -0.65

0.60-0.55

0.50-0.45

0.40-0.35

0.30-025

0 . 2 0 -

0.15

0 . 1 0 -

0.05

0.0 0 -

Figure 3.7 Thin layer chromatogram of petroleum ether fractions obtained from fresh

and dried peels of CA.

F: Fresh peels; D30: Dried peels (30 °C for 4 d); D50: Dried peels (50 °C for 2 d); R f: Retention factor. Developing solvent system: Hex:EtOAc (50:50, v/v). UV light: 254 nm. Amount o f fractions per spot:100//g.


Comparison of the insecticidal activity of the petroleum ether fractions obtained from

fresh and dried peels was done against BO adults using the Petri dish exposure

bioassay. The screening concentration was set at 100 /ig/cm .

64

The results of this experiment are displayed in Table 3.7. The fraction obtained from

fresh peels was highly active (94 % mortality 96 h after treatment), whereas the

fractions obtained from dried peels were inactive (0-4 % mortalities 96 h after

treatment). Moreover, at all mortality recording times, the mortality of the fraction

obtained from fresh peels was significantly greater than the mortality of the two

fractions obtained from dried peels and the control.

Table 3.7 Percentage mortality of BO adults exposed to petroleum ether fractions

obtained from fresh and dried peels of CA at 100 /ig/cm2 using the Petri dish exposure

bioassay.

% MortalityPeel Mean* ± SEb, at hours after treatment

24 48 72 96

Fresh 20 ± 7.1a 84 ± 4.0a 92 ± 2.0a 94 ± 2.4a

Dried 30 °C 0 ± 0.0b 2 ± 2.0b 4 ± 2.4b 4 ± 2.4b

Dried 50 °C 0 ± 0.0b 0 ± 0.0b 0 ± 0.0b 0 ± 0.0b

Control0 0 ± 0.0b 2 ± 2.0b 2 ± 2.0b 2 ± 2.0b

“Mean o f 5 replicates o f 10 insects each (5 males and 5 females). bSE: Standard error o f mean. 'Treated with DCM. Means within a column followed by the same letter are not significantly different (P = 0.05, Tukey’s HSD test). Mortalities were transformed before ANOVA to arcsine Vp, where p is the proportion o f dead insects. Means ± SE o f untransformed data are reported.

These results indicate that the insecticidal active chemicals present in the fraction

obtained from fresh peels were at least partly decomposed during the drying

treatments. Consequently, the proportions of the insecticidally active chemicals in the

fractions obtained from the dried peels, if present, were not so great as to produce any

sign of activity against BO adults.

65

3.3.4 Effect of different extraction methods and solvents on extraction yield,

chemical profile and insecticidal activity of peel fractions and extracts of CA

Peel fractions and extracts of CA were obtained as described in section 3.2.1 in order

to determine the combination of extraction method and initial extraction solvent,

which recovered the insecticidal active chemicals in greater proportions.

Extraction yield

The extraction yields of peel fractions and extracts of CA are presented in Table 3.8.

There were no differences between extraction yields obtained using the same initial

extraction solvent, but different extraction method. Moreover, the extraction yield of

the petroleum ether extracts was more than 1.4 and 6.5 times the yield of MeOH-PE

and H2O-PE fractions, respectively.

Table 3.8 Extraction yield of peel fractions and extracts obtained from 100 g fresh

peels of CA using different extraction methods and initial extraction solvents.

Extractionsolvents8

Extraction method

Conventional solid-liquid Ultrasound

Extraction yield (mg/g fresh)b

PE 3.74 4.02MeOH-PE 2.65 2.60H2O-PE 0.55 0.58

*PE: Initial extraction with petroleum ether; MeOH-PE: Initial extraction with methanol, liquid-liquid partitioning with petroleum ether; H20-PE: Initial extraction with water, liquid-liquid partitioning with petroleum ether. Extraction yield of fractions/extracts was recorded as fractions/extracts weight (mg) per g of fresh peels extracted.

Although the ultrasound extraction method employed less solvent and was much more

rapid compared to the conventional solid-liquid extraction method, similar yields

were obtained by both methods, indicating that the ultrasound extraction method is

66

more efficient than the conventional solid-liquid extraction method. The low yield

obtained by both methods using water for the initial extraction is explained by the

inability of water to extract non-polar chemical compounds. Moreover, the lower

yield obtained by both methods using methanol than petroleum ether for the initial

extraction can be explained by the ability of petroleum ether to extract selectively

non-polar compounds, whereas methanol is extracting compounds representing a wide

polarity range.

Chemical profile

Comparison of the chemical profiles of peel fractions and extracts were performed by

TLC using Hex:EtOAc (50:50, v/v) as the developing solvent system. In Figure 3.8

the analysed petroleum ether peel fractions and extracts of CA are shown. TLC

analysis revealed small quantitative differences between the chemical profiles of

fractions/extracts obtained by using the same initial extraction solvent, but different

extraction methods. Moreover, the chemical profiles of the H2 O-PE fractions obtained

by both extraction methods differ substantially both quantitatively and qualitatively

with the four other chemical profiles. Furthermore, among the two MeOH-PE

fractions and the two PE extracts there were some differences, but not such great as

that stated above.

These findings indicate that the effect of extraction method in the chemical profile of

the fractions/extracts was not crucial. By contrast, the choice of initial extraction

solvent greatly affected the chemical profiles of these fractions/extracts. Moreover,

the inability of water to extract non-polar chemical compounds is obvious. It must be

noticed that although some chemicals of the H2O-PE fractions with R{ below 0.5 were

67

present in greater proportions compared to the other four extracts, the amount of these

chemicals in these fractions was lower as their extraction yields were at least 4.5 and

6.5 times less compared to the MeOH-PE fractions and PE extracts, respectively.

1.00 —

0.95-

0.90 — 0.85 -

0.80 — 0 .75-

0.70 — 0.65

0 .6 0 - 0.55

0 .50 - 0.45

0 .4 0 - 0.35

0 .3 0 - 0.25

0.20 —

0.15 -

0.10 —

0.05 -

0.00 —

Figure 3.8 Thin layer chromatogram of peel fractions and extracts of CA obtained by

different extraction methods and initial extraction solvents.

1: Conventional solid-liquid PE; 2: Ultrasound PE; 3: Conventional solid-liquid MeOH-PE; 4: Ultrasound MeOH-PE; 5: Conventional solid-liquid H20-PE; 6: Ultrasound H20-PE; R {: Retention factor. Developing solvent system: Hex:EtOAc (50:50, v/v). UV light: 254 nm. Amount o f fractions/extracts per spot: 100 fig.


Peel fractions and extracts of CA were evaluated for their insecticidal activity against

BO adults using the Petri dish exposure bioassay. The screening concentration used

68

for the initial experiment was 100 /ig/cm2. Only fractions/extracts that gave significant

activity at this concentration were further tested at 50 /ig/cm2 and moreover, only

active fractions/extracts at 50 //g/cm2 were tested at 25 /*g/cm2.

The results of the three experiments are shown in Table 3.9. The H2O-PE fractions

obtained by the two extraction methods and the MeOH-PE fraction obtained by the

Table 3.9 Percentage mortality of BO adults exposed to peel fractions and extracts of

CA obtained by different extraction methods and initial extraction solvents using the

Petri dish exposure bioassay.

Concentration/^g/cm2

% MortalityFraction/Extract8 Meanb ± SEC, at hours after treatment

24 48 72 96

Conv. S-L PE 42 ± 8.0a 94 ± 4.0a 98 ± 2.0a 100 ± 0.0aConv. S-L MeOH-PE 14 ± 6.8b 62 ± 5.8b 80 ± 5.5b 82 ± 6.6bConv. S-L H20-PE 2 ± 2.0bc 4 ± 2.4c 4 ± 2.4c 4 ± 2.4c

100 Ultrasound PE 52 ± 8.0a 92 ± 3.7a 96 ± 2.4a 100 ± 0.0a

Ultrasound MeOH-PE 0 ± 0.0c 2 ± 2.0c 2 ± 2.0c 2 ± 2.0cUltrasound H20-PE 2 ± 2.0bc 2 ± 2.0c 2 ± 2.0c 2 ± 2.0cControl*1 0 ± 0.0c 2 ± 2.0c 2 ± 2.0c 2 ± 2.0c

Conv. S-L PE 10 ± 4.5a 48 ± 7.3a 84 ± 6.0a 86 ± 5.1a

50Conv. S-L MeOH-PE 2 ± 2.0a 20 ± 7.1b 24 ± 5.1b 24 ± 5.1bUltrasound PE 6 ± 4.0a 50 ± 5.5a 76 ± 6.0a 80 ± 4.5aControl 0 ± 0.0a 2 ± 2.0c 2 ± 2.0c 2 ± 2.0cConv. S-L PE 4 ± 4.0a 24 ± 8.7a 34 ± 7.5a 36 ± 7.5a

25 Ultrasound PE 4 ± 2.4a 18 ± 5.8a 32 ± 7.3a 38 ± 7.3aControl 0 ± 0.0a 0 ± 0.0b 0 ± 0.0b 0 ± 0.0b

*Conv. S-L: Conventional solid-liquid; PE: petroleum ether extract; MeOH-PE: petroleum ether fraction o f methanol extract; H20-PE: petroleum ether fraction o f aqueous extract. bMean o f 5 replicates o f 10 insects each (5 males and 5 females). CSE: Standard error o f mean. dTreated with DCM. The three experiments (100, 50 and 25 /Yg/cm2) were not run simultaneously. Means within a column o f a single concentration followed by the same letter are not significantly different (P = 0.05, Tukey’s HSD test). Mortalities were transformed before ANOVA to arcsine V/?, where p is the proportion o f dead insects. Means ± SE o f untransformed data are reported.

69

ultrasound extraction method were inactive (2-4 % mortalities 96 h after treatment) at

100 /ig/cm2. The mortality caused by the MeOH-PE fraction obtained by the

conventional solid-liquid extraction method was significant lower compared to the PE'y

extracts obtained by both methods at 100 //g/cm at all mortality recording times and

at 50 //g/cm2 at 48, 72 and 96 h after treatment. Moreover, there were no significant

statistical differences between the two PE extracts at the three screening

concentrations at all mortality recording times.

The results of the three experiments indicate that the insecticidal active chemicals

were present in the two PE extracts in similar proportions, but greater compared to the

MeOH-PE fraction obtained by the conventional solid-liquid extraction method. The

H2 O-PE fractions obtained by both extraction methods and the MeOH-PE fraction

obtained by the ultrasound extraction method either did not contain these chemicals or"S

contain them in such concentrations that can not cause mortality even at 100 //g/cm .

3.3.5 Effect of fruit ripeness on extraction yield, chemical profile and insecticidal

activity of petroleum ether peel extracts of CA

Petroleum ether peel extracts were obtained as described in section 3.2.1 in order to

investigate whether there were differences in the insecticidal activity of the petroleum

ether peel extracts obtained from fruits collected at different times in the same

growing season.

Extraction yield

The dry weight (%) of peels obtained from CA fruits at various ripening stages and

the extraction yield of their petroleum ether extracts are displayed in Table 3.10. The

70

dry weight of peels varied from 30.4 to 34.1 %. There were no differences among

extraction yields of the petroleum ether extracts obtained by the extraction of peels of

fruits collected in August, September and October and also for peels of fruits

collected in November, December, January and February. Moreover, the extraction

yield obtained by the extraction of peels on August, September and October was at

least 1.3 times greater than the extraction yield obtained by the extraction of peels on

November, December, January and February, either expressed as milligrams per gram

of fresh peel or milligrams per gram of dried peel extracted.

Table 3.10 Dry weight (%) of peels obtained from CA fruits at various ripening

stages and extraction yield of their petroleum ether extracts.

Collection Fresh amount extracted (g)

Dry weight (%)

Extraction yieldmg/g fresh* mg/g driedb

August 1 0 0 34.1 4.59 13.5

September 1 0 0 33.5 4.70 14.0

October 1 0 0 32.3 4.43 13.7

November 1 0 0 31.7 3.13 9.9

December 1 0 0 30.9 3.21 10.4

January 1 0 0 31.5 3.31 10.5

February 1 0 0 30.4 3.07 1 0 .1

“Extraction yield o f extracts was recorded as extracts weight (mg) per g o f fresh peels extracted. '‘Extraction yield o f extracts was recorded as extracts weight (mg) per g o f dried peels extracted.

These findings indicate that the greater yields of extracts on August, September and

October compared to November, December, January and February, were due, not to

the dry weight extracted in each case, but rather to changes in the chemical

composition of the peel occurring during the third and fourth collection (during the

change in fruit colour from light green to green-yellow).

71

Chemical profile

Comparison of the chemical profiles of the petroleum ether peel extracts were

performed by TLC using HexiEtOAc (50:50, v/v) as the developing solvent system.

In Figure 3.9 the analysed petroleum ether peel extracts of CA are shown. TLC

analysis revealed that below Rf 0.55 the chemical profiles of the seven extracts had

minor quantitative differences. However, the proportions of chemicals with Rf above

0.55 increased during the ripening process. That finding indicates that the chemical

composition of the peels was affected by the ripening process of the fruit.

A S O N D J F

Figure 3.9 Thin layer chromatogram of petroleum ether peel extracts of CA obtained

by fruits at various ripening stages.

A, S, O, N, D, J and F: Petroleum ether peel extracts obtained by fruits collected on August, September, October, November, December, January and February, respectively; R f Retention factor. Developing solvent system: Hex:EtOAc (50:50, v/v). UV light: 254 nm. Amount o f extracts per spot:100/^g.

72


The petroleum ether peel extracts of CA obtained from fruits collected monthly from

August to February were evaluated for their insecticidal activity against BO adults

using the Petri dish exposure bioassay. The screening concentration used for the

initial experiment was 50 //g/cm2. As all extracts were significant active at this

concentration, all extracts were subsequently tested at 25 //g/cm2. Only active extracts

at 25 //g/cm2 were further tested at 12.5 //g/cm2.

The results of the three experiments are shown in Table 3.11. Although there were no

significant statistical differences in mortality among the seven extracts, 96 h after

treatment at the screening concentration of 50 //g/cm2, there was evidence that the

extracts obtained on August, September and October were not as active as the extracts

obtained on November, December, January and February. This was confirmed from

the results of the experiment at the 25 //g/cm2 concentration where the flies mortality

in Petri dishes treated with peel extracts obtained from fruits collected on August,

September and October was significantly lower than that of peel extracts obtained

from fruits collected on November, December, January and February at 72 and 96 h

after treatment. Moreover, the results of the experiment at 12.5 //g/cm2 shown that

there was no significant statistical differences among the extracts obtained on

November, December, January and February at all mortality recording times.

The results indicate that the insecticidal active chemicals were present in greater

proportions in the extracts obtained on November, December, January and February

compared to the extracts obtained on August, September and October. It can be

73

concluded that changes in the chemical composition of the peels occurring during the

third and fourth collection affected the insecticidal activity of the extracts.

Table 3.11 Percentage mortality of BO adults exposed to petroleum ether peel

extracts of CA obtained by fruits at various ripening stages using the Petri dish

exposure bioassay.

Concentration//g/cm2

% Mortality

Extract Mean8 ± SEb, at hours after treatment

24 48 72 96

August 6 ± 2.4a 42 ± 5.8b 82 ± 4.9a 8 8 ± 4.9a

September 6 ± 2.4a 44 ± 5.1b 8 8 ± 3.7a 90 ± 4.5a

October 10 ± 3.2a 52 ± 8 .6 ab 90 ± 4.5a 92 ± 3.7a

50November

December

2 2 ± 8 .0 a

18 ± 4.9a

6 8 ± 6 .6 ab

70 ± ll.Oab

92 ± 5.8a

1 0 0 ± 0 .0 a

98 ± 2.0a

1 0 0 ± 0 .0 a

January 14 ± 7.5a 64 ± 7.5ab 94 ± 4.0a 98 ± 2.0a

February 22 ± 5.8a 78 ± 5.8a 96 ± 2.4a 1 0 0 ± 0 .0 a

Control0 2 ± 2 .0 a 4 ± 2.4c 4 ± 2.4b 4 ± 2.4b

August 0 ± 0 .0 a 18 ± 3.7b 22 ± 4.9b 24 ± 4.0bSeptember 2 ± 2 .0 a 22 ± 5.8ab 28 ± 7.3b 28 ± 7.3b

October 4 ± 2.4a 20 ± 4.5ab 28 ± 3.7b 30 ± 4.5b

25November

December

8 ± 3.7a

6 ± 2.4a

44 ± 8 .lab

56 ± 10.8a

70 ± 4.5a

80 ± 8.9a

72 ± 3.7a

84 ± 8.1a

January 8 ± 3.7a 46 ± 12. lab 74 ± 9.3a 80 ± 7.1a

February 10 ± 3.2a 56 ± 11.2a 82 ± 6 .6 a 8 6 ± 5.1a

Control 0 ± 0 .0 a 0 ± 0 .0 c 0 ± 0 .0 c 0 ± 0 .0 c

November 6 ± 4.0a 32 ± 8 .6 a 34 ± 9.3a 36 ± 8.1a

December 2 ± 2 .0 a 24 ± 8.7a 40 ± 7.1a 42 ± 8.0a

12.5 January 4 ± 2.4a 24 ± 5.1a 28 ± 4.9a 28 ± 4.9a

February 4 ± 2.4a 30 ± 7.1a 36 ± 6 .8 a 36 ± 6 .8 a

Control 0 ± 0 .0 a 0 ± 0 .0 b 0 ± 0 .0 b 0 ± 0 .0 b

*Mean o f 5 replicates o f 10 insects each (5 males and 5 females). bSE: Standard error o f mean. *Treated with DCM. The three experiments (50, 25 and 12.5 //g/cm2) were not run simultaneously. Means within a column o f a single concentration followed by the same letter are not significantly different (P = 0.05, Tukey’s HSD test). Mortalities were transformed before ANOVA to arcsine Vp, where p is the proportion o f dead insects. Means ± SE o f untransformed data are reported.

74

3.4 DISCUSSION

Screening experiments using the Petri dish exposure bioassay against adults of BO

shown that the petroleum ether fraction of the fruit methanol extract was the only

insecticidal active fraction among the different solvent fractions of fruits, leaves and

shoots of CA. TLC analyses revealed that the proportion and range of chemical

substances differ greatly among the fractions (petroleum ether, dichloromethane, ethyl

acetate and residual) of the same plant part and also among the plant parts. The

different insecticidal potency of CA fractions should be accounted to their differences

in chemical composition. The bioactive chemicals are of low polarity since they were

recovered in a non-polar solvent such as petroleum ether. Fractions that did not show

any sign of activity either did not contain these chemicals or contain them in low

concentrations.

Evaluation of the insecticidal activity of the different fruit tissues revealed that the

bioactive chemicals present in the petroleum ether fraction of the fruits were mainly

synthesised and/or accumulated in the peels. TLC analysis revealed that the chemical

composition of fruit tissues differ substantially indicating that chemical substances are

produced and/or accumulated specifically in certain fruit tissues. Citrus peel extracts

obtained with non-polar solvents have been documented to be insecticidally active by

different bioassay methods. Abbassy et al. (1979) showed that lime, navel orange and

grapefruit peel oils obtained using benzene as the extraction solvent exhibited

significant activity against Tribolium confusum and Sitophilus granarius by the Petri

dish exposure bioassay. In addition, Su et al. (1972b) reported that lemon, grapefruit,

lime, kumquat and tangerine peel oils obtained using hexane as the extraction solvent

exhibited significant activity against Sitophilus oryzae and Callosobruchus maculates

75

by the topical application bioassay. It is worth-noted that Su et al. (1972b) had

removed the volatile materials of their extracts by lyophilisation, whereas the extracts

obtained by Abbassy et al. (1979) contained both volatile and non-volatile chemical

substances. In this study, most volatiles have removed from all fractions/extractions

under a stream of nitrogen.

Petroleum ether peel extracts lost their insecticidal activity when peels were oven-

dried prior to the extraction, either at 30 °C for 4 d or 50 °C for 2 d. Moreover, the

extraction yields of the two petroleum ether peel fractions obtained from dried peels

were similar and at least 1.5 times less than that of the petroleum ether fraction

obtained from fresh peels. Considering these results, it can be concluded that the

bioactive chemicals present in the peels were transformed at least partly to less toxic

or non-toxic chemicals due to enzymatic and/or chemical processes or reactions. TLC

analysis revealed only quantitative differences when the two petroleum ether peel

fractions obtained from dried peels were compared to the petroleum ether fraction

obtained from fresh peels. These differences may be accountable for the inactivity of

the two petroleum ether peel fractions obtained from dried peels. Sheppard (1984)

reported that oils from dried peels were less active than oils obtained from fresh. In

addition, Salvatore et al. (2004) found that toxicity against larvae of CC decreased

when diethyl ether extracts of lemon peel were obtained from stored lemons (storage

conditions: 25 ± 2 °C temperature and 60 - 70 % relative humidity). These authors

attributed that decrease in toxicity to the decrease in the concentration of total

coumarins after harvest (35 % over 2 months). Moreover, Thomas and Callaghan

(1999) noted that a crude aqueous lemon peel extract had low persistence as a

larvicide. Activity of this extract against Culex pipiens larvae was reduced even 72 h

76

after the extract had been added to the water. The lack of persistence of lemon peel

extract was related by these authors to the volatility of the active ingredients. In

contrast, Don-Pedro (1985) demonstrated that powdered sun-dried orange and

grapefruit peels applied as food admixtures, exhibited toxic and repellent properties

against Dermestes maculatus and Callosobruchus maculatus. However, Don-Pedro

had not used fresh peels in order to compare the activity between fresh and sun-dried

peels.

Bioassays conducted with peel fractions/extracts obtained using three different initial

extraction solvents and two different extraction methods, revealed that the petroleum

ether extracts obtained by both methods were the most active. Since the ultrasound

extraction method is more rapid and employs less solvent than the conventional solid-

liquid extraction and also it afforded greater extraction yield, this method was

preferred in all subsequent extractions in this thesis. The reduced activity of the

petroleum ether fraction obtained by the conventional solid-liquid extraction method

indicate that petroleum ether extract more selectively the insecticidal chemicals than

methanol. Diethyl ether, ethyl acetate and methanol lemon peel extracts were obtained

by Salvatore et al. (2004) using an ultrasound bath at 20 °C for 20 min. Results of a

larval toxicity bioassay against CC shown that the diethyl ether extract, which contain

mainly non-polar chemicals, was the most toxic, supporting our results.

Fruit maturity affected petroleum ether peel extracts bioactivity. Results indicate that

extracts from ripe fruits, collected on November, December, January and February,

were more active than extracts obtained from unripe fruits, collected on August,

September and October. TLC analysis revealed that the chemical composition of the

77

peels was affected by the ripening process of the fruit. It seems that changes in the

chemical profile of peels during October were responsible for the increased activity of

ripe peel extracts. Citrus fruits are resistant to attack by tephritid fruit flies prior to the

occurrence of peel senescence (Greany et al., 1994). The resistance of lemons to CC

(Back and Pemberton, 1918; Quayle, 1914, 1929) and to Anastrepha suspense

(Greany et a l, 1983) was attributed by these authors, among other reasons, to the peel

oil. More specifically, da Silva Branco et a l (2000) reported that oxygenated

monoterpene aldehydes, like citral, are responsible for the chemical resistance of

lemons to attack by CC. Although unripe fruits are more tolerant to insect attack than

ripe, our results show that insecticidal active chemicals were present in greater

proportion to ripe fruits. That discrepancy can be explained by the presence in our

petroleum ether peel extract of non-volatile compounds mainly, whereas fruit

tolerance is attributed to volatile compounds.

Screening experiments revealed that CA peels possess chemical substances that are

bioactive to BO flies. This finding prompted us to continue our study in two

directions: (a) to evaluate the insecticidal activity of a peel extract by different

bioassays against BO and CC adults, and (b) to isolate the bioactive chemicals of the

active peel extract using bioactivity-guided fractionation procedures. These studies

are reported in Chapters 4 and 5, respectively.

78

CHAPTER 4

EVALUATION OF THE INSECTICIDAL ACTIVITY OF A

PETROLEUM ETHER PEEL EXTRACT OF CITRUS

AURANTIUM TO BACTROCERA OLEAE AND CERA77775

CAPITATA ADULTS

4.1 INTRODUCTION

Insecticide bioassays are experiments done with an insecticide to estimate the

probability that an insect population will respond in the desired manner and so be

made innocuous (Robertson and Preisler, 1992). Initial evaluation of a candidate

insecticide takes place in the laboratory and if it is promising it is further tested in a

series of glasshouse and field trials. Several bioassays methods have been developed

in order to test insecticides in the laboratory.

The most commonly employed method is topical application in which the insecticide

is dissolved in a relatively nontoxic and volatile solvent and applied directly to a

specific part of the outer surface of an insect. Usually combinations of a constant

amount of solvent with varied concentrations of the insecticide are used for this

purpose in order to keep the area of contact, as well as the effect of solvent, constant

(Matsumura, 1975). Different solvents, solvent volumes and sites of placement

[dorsal surface of the thorax (Iwuala et al., 1981), ventral abdomen (Nigg et al.,

1994), legs and head (Scott et al., 1986), left eye of moths (Forrester et al., 1993)]

have been used in topical application bioassays. The topical application bioassay

79

permits quantification of the amount of toxicant received by each individual test

insect, but does not reflect the mechanism by which residual insecticides are acquired

by insects (Hinkle et al., 1985), as most insects physically encounter surface residues

by tarsal contact, not dorsal contact (Stark et al., 1995). The clearest advantage of

topical application is that the doses are independent of insect activity. A disadvantage

is that this method may overcome penetration barriers. It is also tedious and requires

manual dexterity (Moyses and Gfeller, 2001).

The treated surface exposure method is another way of exposing insects to an

insecticide. The insecticide in a solvent is applied to the container or the panel surface

where insects are to walk or rest. The solvent is evaporated by rotating the container

or the panel, so that the insecticide is evenly spread over a known area (Matsumura,

1975). Petri dishes (Studebaker and Kring, 2003), vials (Stark et al., 2004), jars (Hu

and Prokopy, 1998) and filter papers (Obeng-Ofori and Reichmuth, 1997) have been

used in treated surface exposure bioassays. The treated surface exposure bioassay

gives a realistic simulation of the field situation, where insects are exposed to

insecticides mainly by tarsal contact. However, this method does not allow the actual

amount of insecticides acquired by insects to be determined (Choo et al., 2000).

Based on the various types of practical control measures and the insect’s mode of

life, several other bioassay methods have been developed for testing the effectiveness

of insecticides toward various insect species. These are: injection (Jefferies et al.,

1992), dipping (Adan et al., 1996), ingestion (Medina et al., 2004), fumigation

(Konstantopoulou et al., 1992), leaf-dip (Lee et al., 1997) and systemic bioassay

(Siskos, 1997).

80

Assessing of mortality is the most immediate concern when insecticides are tested.

However, other effects on insects such as knockdown (Gbewonyo et al., 1993),

repellency, feeding and oviposition deterrency (Talukder and Howse, 1994), reduced

fertility, fecundity and offspring development (Jimenez-Peydro et al., 1995) could be

of great importance. The toxicity of insecticides to a particular insect is expressed as

LD50/LC50 (median lethal dose/concentration). LD50/LC50 is the statistically derived

dose of an insecticide/concentration of an insecticide in an environmental medium,

expected to kill 50 % of test insects in a given population under a defined set of

conditions (Duffus, 1993).

Intrinsic factors (species and stage specificity, sex, size and resistance), extrinsic

factors (temperature, humidity, illumination, population density and food),

experimental factors and experimental techniques are factors which affect the

evaluation of an insecticide (Matthews, 1997).

In this chapter the insecticidal activity of a petroleum ether peel extract obtained using

the ultrasound extraction method was evaluated against males and females of BO and

CC adults by the topical application and the Petri dish exposure bioassay. In addition,

a fumigation bioassay was employed in order to examine whether the activity of the

extract in the Petri dish exposure bioassay is due to fumigation rather than contact and

moreover to investigate the fumigant toxicity of this extract.

81

4.2 METHODS

4.2.1 Extraction of CA peels

Twenty-five green-yellow fruits of CA weighing 70-120 g were collected from each

of the 10 sampling trees on 1-5 November 1998 in order to obtain sufficient amount

of petroleum ether peel extract for the study of its insecticidal properties. Three-

hundred grams of peel per tree were obtained and a series of 30 extractions were

performed by using the ultrasound extraction method and petroleum ether as

extraction solvent (see section 2.2.1). The extractions were performed using the same

amount of peel (100 g) used in all previous extractions, as the proportions and range

of substances extracted can change when scaling up a process (Houghton and Raman,

1998). The extractable materials of the 30 extractions were combined and the

extraction yield of the total extract was estimated by concentrating to near-dryness

1 % of the total extract in the rotary evaporator at 35 °C. The concentrated extract was

subsequently transferred to a 15-ml glass vial, immersed in a water bath at 35 °C and

concentrated to constant weight under a stream of nitrogen. The remaining extract was

concentrated to dryness and stored in the dark at -20 °C until used. The weighing was

performed on the Mettler Toledo balance.

4.2.2 Bioassays

The insecticidal properties of the petroleum ether peel extract were evaluated using

the Petri dish exposure, fumigation and topical application bioassays (see section

2.2.2) against adults of BO and CC. Stock solutions of the petroleum ether peel

extract were prepared in DCM, as were subsequent dilutions. Fresh solutions were

prepared for each replicate. Concentrations/doses for each replicate were applied from

the lowest to the highest.

82

In the Petri dish exposure and topical application bioassays, males and females of BO

and CC were tested separately, but simultaneously. For each sex, 5 replicates of 10

flies each were employed for the 5 concentrations/doses and the control. The 5

replicates were conducted using 5 different generations of flies. As the main aim of

the experiments performed with the Petri dish exposure and the topical application

bioassays was the precise estimation of LC50S/LD50S, respectively five

concentrations/doses were chosen for each sex of the two test insects on the basis of

preliminary range-finding studies performed by the two biological evaluation

techniques, in order to produce mortalities evenly distributed between 25 and 75 % at

the final day of recording of mortality (Robertson et al., 1984).

Petri dish exposure bioassay

The concentrations tested against both sexes of BO were 10, 15, 20, 25 and 30

/vg/cm2. The concentrations employed against males of CC were 50, 60, 70, 80 and 90

//g/cm2, while the concentrations used against females of CC were 100, 120, 140, 160

and 180 //g/cm2. Mortality for males and females of BO was recorded every 24 h for 4

d and for males and females of CC every 24 h for 3 d, after which no further increase

in mortality to both insect species was recorded.

Topical application bioassay

The doses employed against both sexes of BO and the males of CC were 20, 30, 40,

50 and 60 //g/insect, while the doses employed against females of CC were 30, 50, 70,

90 and 110 //g/insect. BO and CC mortalities of both sexes were recorded every 24 h

for 6 and 4 d, respectively after which there was no further increase in mortality to

both species. Ten adults of each sex of the two test insects were randomly selected at

83

the end of each replicate and their body weight was measured. The mean body weight

for male and female flies of BO and CC was determined and so doses can be

expressed, not only as //g/insect, but also as micrograms per milligram adult body

weight (//g/mg).

Fumigation bioassay

Males and females of BO were equally susceptible when exposed to the petroleum

ether peel extract using the Petri dish exposure bioassay. Considering that result,

males and females were tested together (5 males and 5 females/Petri dish) in the

fumigation bioassay. The concentration used was 4mg/Petri dish, which is

approximately 2-fold greater than the highest concentration used against BO in the

concentration-response Petri dish exposure bioassay.

Males were significantly more susceptible than females of CC in the Petri dish

exposure bioassay. Consequently, males and females were tested in separate

experiments (10 males or females/Petri dish). The concentration employed against

males was 12 mg/Petri dish, while the concentration employed against females was 24

mg/Petri dish. These concentrations were approximately 2-fold greater respectively

than the highest concentration used against males and females of CC in the

concentration-response Petri dish exposure bioassay.

Four Petri dishes for each concentration and their corresponding controls (see section

2.2.2), were replicated 5 times using flies of different generations. Mortality of BO

adults was recorded every 24 h for 4 d, whereas mortality of CC adults (both sexes)

every 24 h for 3 d.

84

4.3 RESULTS

4.3.1 Petri dish exposure bioassays

Petri dish exposure bioassays were performed in order to compare the susceptibility of

BO and CC adults to the petroleum ether peel extract of CA. Moreover, as males and

females of both species were tested separately the susceptibility of sexes was also

compared.

BO

The results obtained by the Petri dish exposure bioassay against males and females of

BO are presented in Figure 4.1 and Figure 4.2, respectively. At 24 h after treatment,

mortality of males and females of BO was not significantly different from the control

even at the highest concentration of the petroleum ether peel extract tested (30

//g/cm ). However, at 48, 72 and 96 h after treatment, mortality of males and females

was significantly higher compared to the control at concentrations of 15, 20, 25 and

30 //g/cm of the petroleum ether peel extract. Mortality of males and females caused

by the concentration of 1 0 /ig/cm2 was greater than the control, but not significantly

different. The mortality results reveal that mortality in males and females of BO

increased with increasing concentration, indicating that mortality is concentration-

dependent.

A summary of probit analyses of mortality data for males and females of BO exposed

to different concentrations of the petroleum ether peel extract of CA using the Petri

dish exposure bioassay is presented in Table 4.1. Regression lines obtained from

probit analyses are shown in Figure 4.3. The concentrations selected to estimate the

LC50 of males and females at the end-point mortality (96 h after treatment) cause low

85

■aj*oE4» 0D 08§Ou,«aBss

100

90

80

70

60

50

40

30

20

10

0 nilI

24

a

i i

48 72 96

ooON

Hours after treatment

Figure 4.1 Percentage mortality of BO male adults exposed to different concentrations (pg/cm ) of the petroleum ether peel extract of CA using

the Petri dish exposure bioassay.

Columns represent mean o f 5 replicates o f 10 males each. Bars at each data point indicate the standard error. Columns o f the same mortality recording time labelled by the same letter are not significantly different (P = 0.05, Tukey’s HSD test). Mortalities were transformed before ANOVA to arcsine Vp, where p is the proportion o f dead insects. Means and standard errors o f untransformed data are presented.

1

100 n

48t,OEa>M)08

aQi<JU4)ac8s

90

80

70 H

60

50

40

30

20

10

024 48 72 96

oo<i


2Figure 4.2 Percentage mortality of BO female adults exposed to different concentrations ( ju g /c m ) of the petroleum ether peel extract of CA

using the Petri dish exposure bioassay.

Columns represent mean o f 5 replicates o f 10 females each. Bars at each data point indicate the standard error. Columns o f the same mortality recording time labelled by the same letter are not significantly different (P = 0.05, Tukey’s HSD test). Mortalities were transformed before ANOVA to arcsine Vp, where p is the proportion o f dead insects. Means and standard errors o f untransformed data are presented.

9

99999999


exposed to different concentrations of the petroleum ether peel extract of CA using


h Sex n LC50(^g/crn2) 95 % CL Slope ± SE Intercept ± SE / df P

48 (5 250 23.0 2 0 . 8 - 26.2 3.76 ±0.57 -0.13 ±0.75 24.8 23 0.35848 ? 250 23.8 21.3-27.5 3.60 ±0.57 0.04 ± 0.75 27.7 23 0.225

72 s 250 19.1 17.6-20.8 4.88 ± 0.60 -1.25 ±0.78 27.0 23 0.257

72 ? 250 18.1 16.6- 19.7 4.77 ±0.58 -1.00 ±0.75 20.7 23 0.599

96 (5 250 18.8 17.3-20.4 5.02 ±0.60 -1.40 ±0.78 26.5 23 0.280

96 ? 250 17.8 16.4 - 19.3 5.10 ±0.60 -1.38 ±0.77 19.2 23 0.691

Column headings are abbreviated as follows: h is the hours after treatment; n is the numbers o f insects tested excluding the controls; 95 % CL is the 95 % confidence limits; Slope ± SE is the slope ± standard error; Intercept ± SE is the intercept ± standard error; x 2 is the Pearson’s x 2 goodness-of-fit test; df is the degrees o f freedom; P is the probability.

6.5

oE

- Males Females

20.5 1.5

6.5

MalesFemales

0.5 1.5 21Log (jig/cm2) Log (ng/cm2)

6.5

oE.o©uA*

MaksFemales

0.5 1.5 2

Log (fig/cm2)

Figure 4.3 Log (/vg/cm2) - probit mortality lines calculated by probit analysis for

males and females of BO exposed to the petroleum ether peel extract of CA using the

Petri dish exposure bioassay at 48 (A), 72 (B) and 96 (C) h after treatment.

88

i

mortality (14 % or less) at 24 h after treatment; consequently LC50S were not

estimated at this time. All P-values of Table 4.1 were much greater than 0.05 and so

Pearson’s x2 goodness-of-fit tests were not significant, indicating that the data fit the

assumptions of the probit model adequately.

The LC50 estimated from probit analysis for males at 48 h after treatment, was slightly

lower (23 fig/cm2) than females (23.8 fig/cm2), whereas the LC5 0S for males at 72 and

96 h after treatment (19.1 and 18.8 fig/cm2, respectively) were slightly higher than for

females at the same mortality recording time (18.1 and 17.8 fig/cm2, respectively).

However, the 95 % confidence limits of males and females overlapped at all mortality

recording times, indicating that males and females of BO were equally susceptible to

the petroleum ether peel extract of CA. Moreover, the action of the petroleum ether

peel extract against males and females of BO was the same, as the slopes of the

regression lines were not significantly different at all mortality recording times.

CC

The results obtained by the Petri dish exposure bioassay against male adults of CC are

presented in Figure 4.4. At 24 h after treatment, only the mortality caused by the

concentration of 90 jug/cm2 was significantly greater than that of the control, although

10 % mortality was caused by the concentrations of 60, 70 and 80 jug/cm2. At 48 and

72 h after treatment, mortalities caused by all concentration tested were significantly

greater than that of the control.

89

os 60

be abc

ab ab

24 48 72

\o©


Figure 4.4 Percentage mortality of CC male adults exposed to different concentrations (/^g/cm ) of the petroleum ether peel extract of CA using


Columns represent mean o f 5 replicates o f 10 males each. Bars at each data point indicate the standard error. Columns o f the same mortality recording time labelled by the same letter are not significantly different (P = 0.05, Tukey’s HSD test). Mortalities were transformed before ANOVA to arcsine V/?, where p is the proportion o f dead insects. Means and standard errors o f untransformed data are presented.

Mortality trends of female adults of CC (Figure 4.5) were quite similar to those of

male adults of CC, although the concentrations tested against females were 2-fold

greater than that tested against females. 24 h post-treatment, even the highest

concentration tested (180 //g/cm2) which caused 14 % mortality was not significantly

different from the control. At 48 and 72 h after treatment, mortality levels caused by

the concentrations of 120, 140, 160 and 180 ywg/cm2 were significantly greater than

that of the control, while the mortality ( 1 0 %) caused by the lowest concentration

tested (100 /^g/cm2) was not significantly different from the control. Greater

concentrations caused greater male and female mortalities, showing a clear

concentration-response relationship.

A summary of probit analyses of mortality data for males and females of CC exposed

to different concentrations of the petroleum ether peel extract of CA using the Petri

dish exposure bioassay is presented in Table 4.2. Regression lines obtained from the

probit analyses are shown in Figure 4.6. All the concentrations selected to estimate

the LC50 of males and females at the end-point mortality (72 h after treatment) did not

produce significant mortality levels during the first 24 h. For that reason, LC50S were

not estimated at this time. All / ’-values of Table 4.2 were much greater than 0.05 and

so Pearson’s goodness-of-fit tests were not significant, indicating that the data fit

the assumptions of the probit model adequately.

91

□ 0 DD 100

120

m 140

□ 160

0 180

24 48 72

to


2Figure 4.5 Percentage mortality of CC female adults exposed to different concentrations (jug/cm ) of the petroleum ether peel extract of CA


Columns represent mean o f 5 replicates o f 10 females each. Bars at each data point indicate the standard error. Columns o f the same mortality recording time labelled by the same letter are not significantly different (P = 0.05, Tukey’s HSD test). Mortalities were transformed before ANOVA to arcsine V/?, where p is the proportion o f dead insects. Means and standard errors o f untransformed data are presented.


exposed to different concentrations of the petroleum ether peel extract of CA using


h Sex n LC50(//g/cm2) 95 % CL Slope ± SE Intercept ± SE / df P

48 6 250 73.6 68.4 - 80.5 5.18 ±0.96 -4.67 ± 1.76 20.4 23 0.61748 9 250 149.0 140.9- 159.1 6.93 ± 1.03 -10.07 ±2.21 24.4 23 0.381

72 6 250 72.6 67.3 - 79.5 4.95 ± 0.95 -4.21 ± 1.74 2 2 . 2 23 0.510

72 9 250 147.1 139.2 - 156.8 7.00 ± 1.03 -10.17 ±2.20 23.8 23 0.418

Column headings are abbreviated as follows: h is the hours after treatment; n is the numbers o f insects tested excluding the controls; 95 % CL is the 95 % confidence limits; Slope ± SE is the slope ± standard error; Intercept ± SE is the intercept ± standard error; x 2 is the Pearson’s X2 goodness-of-fit test; df is the degrees o f freedom; P is the probability.

6

I* 5 5 © ̂E.ts 4.5JO

£ 4

3.51.5

’ Males Females

2.5

Log (jig/cm2)

6

I* 5 5 "* © Ê.ts 4.5 .c£ 4

3.5

B

1.5

MalesFemales

2.5

Log (Mg/cm2)

Figure 4.6 Log (//g/cm ) - probit mortality lines calculated by probit analysis for

males and females of CC exposed to the petroleum ether peel extract of CA using the

Petri dish exposure bioassay at 48 (A) and 72 (B) h after treatment.

The LC50S estimated from probit analysis for females of CC at 48 and 72 h after

treatment (149.0 and 147.1 //g/cm2, respectively), were 2-fold greater than that of

males (73.6 and 72.6 //g/cm2, respectively). At both mortality recording times, males

of CC were significantly more susceptible to the exposure of the petroleum ether peel

extract of CA than females as their 95 % confidence limits did not overlap.

93

At the end-point mortality of BO and CC (96 and 72 h after treatment, respectively)

the LC50 of CC males was 3.9 times greater than that of BO males. Moreover, the

LC50 of CC females was 8.3 times greater than that of BO females. That difference

(more than 2 -fold) in the degree of susceptibility between males and females

accounted to the difference in susceptibility between CC males and females. These

results indicate that at the end-point mortality, BO males and females were

significantly more susceptible when exposed to the petroleum ether peel extract of CA

using the Petri dish exposure bioassay than CC males and females, respectively.

Moreover, CC adults reached the end-point mortality one day earlier than BO adults,

indicating that the extract acted more rapidly against CC adults.

4.3.2 Fumigant bioassays

In the following bioassays, the possibility that mortality caused by the petroleum ether

peel extract of CA, in the Petri dish exposure bioassay, was due to fumigation rather

than contact and the fumigant action of the petroleum ether peel extract was

investigated.

BO

The results obtained by the fumigation bioassay against BO adults (males and females

were tested together) are presented in Table 4.3. At all mortality recording times only

the two treatments in which adults of BO were exposed to the petroleum ether peel

extract (either applied to the total inner surface or only to the bottom of the Petri

dishes) by contact, caused significant mortality. This result reveals the absence of

fumigant action in the Petri dish exposure bioassays and indicates that mortality

observed was the result of contact o f BO adults with the petroleum ether peel extract

94

of CA. Moreover, as no significant mortality was observed when BO adults were

exposed only to the fumes of the petroleum ether peel extract in Petri dishes sealed

with Parafilm, a lack of fumigant activity of the petroleum ether peel extract against

BO adults even at a concentration of 157 //g/cm3, was revealed.

Table 4.3 Percentage mortality of BO adults exposed to the petroleum ether peel

extract of CA at 4mg/Petri dish using the fumigation bioassay.

% MortalityTreatment8 Meanb ± SEC, at hours after treatment

24 48 72 96

PI 26 ± 6.0a 74 ± 6.8a 100 ± 0.0a 100 ± 0.0aP2 32 ± 8.6a 76 ± 7.5a 96 ± 2.4a 100 ± 0.0a

P3 0 ± 0.0b 2 ± 2.0b 4 ± 2.4b 4 ± 2.4bP4 2 ± 2.0b 4 ± 2.4b 4 ± 2.4b 4 ± 2.4b

Cl 0 ± 0.0b 0 ± 0.0b 0 ± 0.0b 0 ± 0.0b

C2 0 ± 0.0b 0 ± 0.0b 2 ± 2.0b 2 ± 2.0bC3 0 ± 0.0b 2 ± 2.0b 2 ± 2.0b 2 ± 2.0bC4 2 ± 2.0b 4 ± 2.4b 4 ± 2.4b 4 ± 2.4b

“PI: The extract was applied to the total inner surface of the Petri dishes; P2: The extract was applied only to the bottom of the Petri dishes; P3: The extract was applied only to the bottom of the Petri dishes and a sheet of gauze was fitted inside them; P4: The extract was applied only to the bottom of the Petri dishes, a sheet of gauze was fitted inside them and sealed with Parafilm; Cl, C2, C3 and C4: Controls corresponding to PI, P2, P3 and P4, respectively which were treated with DCM only. bMean of 5 replicates of 10 insects each (5 males and 5 females). CSE: Standard error of mean. Means within a column followed by the same letter are not significantly different (P = 0.05, Tukey’s HSD test). Mortalities were transformed before ANOVA to arcsine V/?, where p is the proportion of dead insects. Means ± SE of untransformed data are reported.

CC

Similar results were obtained when the same bioassay was conducted against CC

(males and females were tested separately) (Table 4.4). Only the two treatments, in

which males and females of CC were allowed to contact the petroleum ether peel

extract, produced significant mortality at all mortality recording times. This result

Table 4.4 Percentage mortality of CC male and female adults exposed to the petroleum ether peel extract of CA at 12mg/Petri dish and 24

mg/Petri dish, respectively using the fumigation bioassay.

Treatment8 -

% MortalityMeanb ± SEC, at hours after treatment

24 48 72 24 48 72

Males Females

PI 40 ± 6.3a 100 ± 0.0a 100 ± 0.0a 38 ± 8.6a 96 ± 4.0a 98 ± 2.0aP2 28 ± 6.6a 94 ± 4.0a 96 ± 2.4a 32 ± 5.8a 92 ± 3.7a 94 ± 4.0aP3 0 ± 0.0b 0 ± 0.0b 0 ± 0.0b 0 ± 0.0b 2 ± 2.0b 2 ± 2.0bP4 2 ± 2.0b 2 ± 2.0b 2 ± 2.0b 0 ± 0.0b 0 ± 0.0b 0 ± 0.0b

Cl 2 ± 2.0b 2 ± 2.0b 2 ± 2.0b 0 ± 0.0b 2 ± 2.0b 2 ± 2.0b

C2 0 ± 0.0b 0 ± 0.0b 0 ± 0.0b 0 ± 0.0b 0 ± 0.0b 0 ± 0.0b

C3 0 ± 0.0b 2 ± 2.0b 2 ± 2.0b 2 ± 2.0b 2 ± 2.0b 2 ± 2.0b

C4 0 ± 0.0b 0 ± 0.0b 0 ± 0.0b 2 ± 2.0b 2 ± 2.0b 2 ± 2.0b

“PI: The extract was applied to the total inner surface of the Petri dishes; P2: The extract was applied only to the bottom of the Petri dishes; P3: The extract was applied only to the bottom of the Petri dishes and a sheet of gauze was fitted inside them; P4: The extract was applied only to the bottom of the Petri dishes, a sheet of gauze was fitted inside them and sealed with Parafilm; Cl, C2, C3 and C4: Controls corresponding to PI, P2, P3 and P4, respectively which were treated with DCM only. '’Mean of 5 replicates of 10 adults of the same sex each. CSE: Standard error of mean. Means within a column followed by the same letter are not significantly different (P = 0.05, Tukey’s HSD test). Mortalities were transformed before ANOVA to arcsine l̂p, where p is the proportion of dead insects. Means ± SE of untransformed data are reported.

reveals the absence of fumigant action in the Petri dish exposure bioassays and

indicates that mortality of CC males and females, as in the case of BO, was due to

contact. Moreover, no mortality was recorded when CC males and females were

exposed only to the fumes of the petroleum ether peel extract in Petri dishes sealed

with Parafilm, indicating a lack of fumigant activity of the petroleum ether peel

extract against CC males and females even at a concentration of 471 and 942 //g/cm3,

respectively.

4.3.3 Topical application bioassays

Topical application bioassays were performed in order to compare the susceptibility

o f BO and CC adults to the petroleum ether peel extract of CA. Moreover, as males

and females of both species were tested separately the susceptibility of sexes was also

compared.

BO

The body weight (mean ± standard error of mean) of BO males and females was 6.0 ±

0.14 and 7.7 ±0 .17 mg, respectively. For completeness, as there was variation in

weight between males and females of BO, doses tested are expressed not only as //g

per insect but also as ng per mg of adult body weight. When doses were expressed as

//g/insect, the same doses were employed against males and females of BO (20, 30,

40, 50 and 60). In contrast, when doses were corrected for the difference in body

weight between males and females, the doses employed against males of BO were

3.34, 5.01, 6 .6 8 , 8.35 and 10.02 //g/mg, while the doses employed against females of

BO were 2.60, 3.90, 5.20, 6.50 and 7.80 //g/mg. The slightly lower doses (//g/mg)

employed against females was a result of their greater body weight.

97

The results obtained by the topical application bioassay against male adults of BO are

presented in Figure 4.7. At 24 h after treatment, even mortality caused by the highest

dose tested (60 //g/insect) was not significantly different from the control. 48 h post

treatment, only the two higher doses tested (50 and 60 //g/insect) caused significantly

greater mortality than that of the control. Doses of 30, 40, 50 and 60 //g/insect caused

significantly greater mortality than that of the control at 72, 96, 120 and 144 h after

treatment, whereas the lowest dose tested ( 2 0 //g/insect) was significantly greater from

the control only 72 and 144 h post-treatment.

Mortality trends of female adults of BO (Figure 4.8) were quite similar to those of

male adults. At 24 and 48 h after treatment, even mortality caused by the highest dose

tested (60 //g/insect) was not significantly different from the control. However, 72, 96,

120 and 144 h after treatment, doses of 30, 40, 50 and 60 //g/insect caused

significantly greater mortality than the control, while the lowest dose tested ( 2 0

//g/insect) was significantly different from the control only at 1 2 0 h after treatment.

The results presented in Figure 4.7 and Figure 4.8 show that mortality of BO males

and females increased with increasing dose, indicating a dose-response effect.

98

100 -1

24 48 72 96 120 144Hours after treatment

Figure 4.7 Percentage mortality of BO male adults treated with different doses (//g/insect) of the petroleum ether peel extract of CA using the

topical application bioassay.

Columns represent mean of 5 replicates of 1 0 males each. Bars at each data point indicate the standard error. Columns of the same mortality recording time labelled by the same letter are not significantly different (P = 0.05, Tukey’s HSD test). Mortalities were transformed before ANOVA to arcsine Vp, where p is the proportion of dead insects.

^ Means and standard errors of untransformed data are presented.

100 -1

90 -

80 -

£ 70 -"flS*©s 60 -©6008 50 -Sw£©a 40 -sS£ 30 -

2 0 -

1 0 -

0 -a a a

24

Ll

48 72 96Hours after treatment

120 144

oo

Figure 4.8 Percentage mortality of BO female adults treated with different doses (//g/insect) of the petroleum ether peel extract of CA using the


Columns represent mean of 5 replicates of 10 females each. Bars at each data point indicate the standard error. Columns of the same mortality recording time labelled by the same letter are not significantly different (P = 0.05, Tukey’s HSD test). Mortalities were transformed before ANOVA to arcsine Vp, where p is the proportion of dead insects. Means and standard errors of untransformed data are presented.

A summary of probit analyses of mortality data for males and females of BO treated

with different doses (expressed as //g/insect and //g/mg) of the petroleum ether peel

extract of CA, using the topical application bioassay, is presented in Table 4.5 and

Table 4.6, respectively. Regression lines [probit mortality versus log (//g/insect) and

probit mortality versus log (//g/mg)] obtained from the probit analyses are shown in

Figure 4.9 and Figure 4.10, respectively. The doses selected to estimate the LD50 of

males and females at the end-point mortality (144 h after treatment) did not cause

significant mortality at 24 and 48 h after treatment and so LD50S were not estimated at

these times. All P-values of Table 4.5 and Table 4.6 were greater than 0.05 and so

Pearson’s y? goodness-of-fit tests were not significant indicating that the data fit the

assumptions of the probit model adequately.

Table 4.5 Summary o f probit analyses of mortality data for males and females of BO

treated with different doses (//g/insect) of the petroleum ether peel extract of CA using

the topical application bioassay.

h Sex n LD50(//g/insect) 95 % CL Slope ± SE Intercept ± SE t df P

72 6 250 59.6 50.2 - 82.4 2.67 ±0.55 0.26 ± 0.88 21.5 23 0.55172 9 250 56.4 48.4 - 73.6 2.87 ±0.56 -0.02 ± 0.90 32.8 23 0.084

96 6 250 44.8 40.0-51.5 3.42 ±0.54 -0.65 ± 0.87 22.7 23 0.479

96 9 250 40.1 35.7 -45.4 3.34 ±0.53 -0.35 ± 0.85 30.7 23 0.130

120 6 250 41.9 37.5 -47.6 3.43 ±0.54 -0.56 ±0.85 21.8 23 0.532

120 9 250 34.6 30.5 - 38.6 3.46 ±0.53 -0.33 ± 0.83 17.3 23 0.793

144 $ 250 40.2 36.0-45.3 3.48 ±0.53 -0.59 ±0.85 19.8 23 0.652

144 9 250 33.0 29.1 -36.8 3.57 ±0.53 -0.42 ± 0.83 23.8 23 0.416

Column headings are abbreviated as follows: h is the hours after treatment; n is the numbers of insects tested excluding the controls; 95 % CL is the 95 % confidence limits; Slope ± SE is the slope ± standard error; Intercept ± SE is the intercept ± standard error; y2 is the Pearson’s y? goodness-of-fit test; df is the degrees of freedom; P is the probability.

101


treated with different doses (//g/mg) of the petroleum ether peel extract of CA using


h Sex n LD50(//g/mg) 95 % CL Slope ± SE Intercept ± SE 7? df P

72 (5 250 9.9 8.4-13.8 2.67 ±0.55 2.33 ± 0.46 21.5 23 0.55172 9 250 7.3 6.3 - 9.6 2.87 ±0.56 2.52 ±0.41 32.8 23 0.084

96 250 7.5 6.7 - 8 . 6 3.42 ± 0.54 2.01 ±0.45 22.7 23 0.479

96 9 250 5.2 4.6 - 5.9 3.34 ± 0.53 2.61 ±0.38 30.7 23 0.130

1 2 0 s 250 7.0 6.3 - 8.0 3.43 ±0.54 2.10 ±0.44 2 1 . 8 23 0.532

1 2 0 9 250 4.5 4.0 - 5.0 3.46 ±0.53 2.74 ± 0.37 17.3 23 0.793

144 (5 250 6.7 6.0 - 7.6 3.48 ±0.53 2.12 ±0.44 19.8 23 0.652

144 9 250 4.3 3.8-4.8 3.57 ±0.53 2.74 ± 0.37 23.8 23 0.416

Column headings are abbreviated as follows: h is the hours after treatment; n is the numbers of insects tested excluding the controls; 95 % CL is the 95 % confidence limits; Slope ± SE is the slope ± standard error; Intercept ± SE is the intercept ± standard error; x2 is the Pearson’s x2 goodness-of-fit test; df is the degrees of freedom; P is the probability.

Except from the different estimates of LD5 0S and their corresponding 95 % confidence

limits when doses were expressed as //g/insect and //g/mg adult body weight, the only

other difference between Table 4.5 and Table 4.6 was the intercept and its standard

error. At all mortality recording times, the LD5 0S estimated for males of BO were

greater than that of females, either doses expressed as //g/insect or //g/mg of adult

body weight. When doses expressed as //g/insect, the 95 % confidence limits

overlapped at all mortality recording. In contrast, when doses were expressed as

//g/mg of adult body weight, the 95 % confidence limits did not overlap at 96, 120 and

144 h after treatment, whereas they overlapped at 72 h after treatment. That results

indicate that males and females of BO did not differ significantly in susceptibility

when doses were expressed as //g/insect, whereas females were more susceptible than

males when doses were expressed as //g/mg adult body weight. Moreover, the action

of the petroleum ether peel extract against BO males and females is similar as the

102

slopes of the regression lines did not differ significantly, in all mortality recording

times.

6.5

5.5

MalesFemales3.5

1 1.5 2

6.5

•O

£ MalesFemales

1.5 21Log (iig/insect) Log (jig/insect)

6.5 6

5.5 5

4.5 4

3.5 3

- Males Females

1.5 21

6.5

5.5

0 J

1 ** 3.5

MalesFemales

1 21.5

Log (fig/insect) Log (jig/insect)

Figure 4.9 Log (/2g/insect) - probit mortality lines calculated by probit analysis for

males and females of BO treated with the petroleum ether peel extract of CA using the

topical application bioassay at 72 (A), 96 (B), 120 (C) and 144 (D) h after treatment.

103

6.5 6

5.5 5

4.5 4

3.5 3

MalesFemales

0 0.5 1.51Log (Hg/mg)

6.5 6

5.5 5

4.5MalesFemales

43.5

31.50 0.5 1

Log (pg/mg)

©eeu - Males

Females* 3.5

0 0.5 1.51

6.5 6

5.5 5

4.5 4

3.5 3

MalesFemales

0 0.5 1 1.5

Log (jtg/mg) Log (pg/mg)

Figure 4.10 Log (//g/mg) - probit mortality lines calculated by probit analysis for

males and females of BO treated with the petroleum ether peel extract of CA using the


CC

The body weight (mean ± standard error of mean) of CC males and females was 6.9 ±

0.16 and 8.0 ± 0.25 mg, respectively. For completeness, as there was variation in

weight between males and females of CC, doses tested are expressed not only as //g

per insect but also as //g per mg of adult body weight. The doses (//g/insect) tested

were 20, 30, 40, 50 and 60 against males and 30, 50, 70, 90 and 110 against females

of CC. However, when doses were expressed as //g/mg, the doses tested were 2.90,

4.35, 5.80, 7.25 and 8.70 against males and 3.75 6.25, 8.75, 11.25 and 13.75 against

females of CC. As a result of the greater body weight of females, the difference

104

between the doses tested against males and females was lower when doses were

expressed as //g/mg.

The results obtained by the topical application bioassay against male and female

adults of CC are presented in Figure 4.11 and Figure 4.12, respectively. At 24 and 48

h after treatment, only the male mortalities caused by the two lower doses (20 and 30

//g/insect) were not significantly different from the control. However, at 72 and 96 h

after treatment only the lowest dose tested (20 //g/insect) against males did not cause

significantly greater mortality than that of control. In contrast, at all mortality

recording times all doses tested against females of CC caused mortalities significantly

greater from the control. The results presented in Figure 4.11 and Figure 4.12 reveal

the existence of a dose-dependent effect on mortality of males and females of CC.

105

H—‘oo

« 60

bed abc

bed abc

24 48 72

Hours after treatment96

Figure 4.11 Percentage mortality of CC male adults treated with different doses (//g/insect) of the petroleum ether peel extract of CA using the


Columns represent mean of 5 replicates of 10 males each. Bars at each data point indicate the standard error. Columns of the same mortality recording time labelled by the same letter are not significantly different (P = 0.05, Tukey’s HSD test). Mortalities were transformed before ANOVA to arcsine Vp, where p is the proportion of dead insects. Means and standard errors of untransformed data are presented.

ab abab ab«8 60

□ 0 30 50 70

□ 90m no

a> 50

9 30

24 48 72Hours after treatment

96

o

Figure 4.12 Percentage mortality of CC female adults treated with different doses (//g/insect) of the petroleum ether peel extract of CA using the


Columns represent mean of 5 replicates of 10 females each. Bars at each data point indicate the standard error. Columns of the same mortality recording time labelled by the same letter are not significantly different (P = 0.05, Tukey’s HSD test). Mortalities were transformed before ANOVA to arcsine where p is the proportion of dead insects. Means and standard errors of untransformed data are presented.

999999999999

A summary of probit analyses of mortality data for males and females of CC treated

with different doses (expressed as //g/insect and //g/mg) of the petroleum ether peel

extract of CA using the topical application bioassay are presented in Table 4.7 and

Table 4.8, respectively. Regression lines [probit mortality versus log (pg/insect) and

probit mortality versus log (//g/mg)] obtained from the probit analyses are shown in

Figure 4.13 and Figure 4.14, respectively. Except from the different estimates of

LD5 0S and their corresponding 95 % confidence limits when doses were expressed as

//g/insect and //g/mg adult body weight, the only other difference between Table 4.7

and Table 4.8 was the intercept and its standard error. The P-value at 48 h after

treatment for male adults of CC was 0.039 (significant) and so the data did not fit the

probit model adequately. In this case a heterogeneity factor was automatically used in

the calculations of confidence limits by SPSS. All other P-values of Table 4.7 and


treated with different doses (//g/insect) o f the petroleum ether peel extract of CA using


h Sex n LD50(//g/insect) 95 % CL Slope ± SE Intercept ± SE / df P

24 3 250 57.9 50.7 - 72.3 3.52 ±0.62 -1.20 ± 1.01 33.5 23 0.07224 9 250 117.6 88.2 - 258.0 1.54 ±0.43 1.82 ±0.79 20.4 23 0.61748 6 250 47.4 41.3-57.7 3.77 ±0.58 -1.32 ±0.93 36.3 23 0.039

48 9 250 75.8 61.1 - 103.1 1.71 ±0.42 1.78 ±0.76 22.2 23 0.508

72 S 250 40.0 36.0-44.6 3.77 ±0.55 -1.03 ±0.87 29.7 23 0.15872 9 250 69.7 57.7 - 86.7 2.01 ±0.42 1.29 ±0.77 20.7 23 0.598

96 3 250 38.8 34.7-43.6 3.52 ±0.53 -0.59 ±0.85 27.4 23 0.240

96 9 250 67.8 56.6 - 82.4 2.15 ±0.42 1.07 ±0.77 20.8 23 0.594

Column headings are abbreviated as follows: h is the hours after treatment; n is the numbers of insects tested excluding the controls; 95 % CL is the 95 % confidence limits; Slope ± SE is the slope ± standard error; Intercept ± SE is the intercept ± standard error; ^ is the Pearson’s x2 goodness-of-fit test; df is the degrees of freedom; P is the probability.

108


treated with different doses (//g/mg) of the petroleum ether peel extract of CA using


h Sex n LD50

(//g/mg) 95 % CL Slope ± SE Intercept ± SE jc2 df P

24 s 250 8.4 7.3 - 10.5 3.52 ±0.62 1.75 ±0.49 33.5 23 0.07224 9 250 14.7 11.0-32.3 1.54 ±0.43 3.20 ±0.41 20.4 23 0.617

48 6 250 6.9 6.0 - 8.4 3.77 ±0.58 1.84 ±0.45 36.3 23 0.039

48 9 250 9.5 7.6 - 12.9 1.71 ±0.42 3.33 ±0.39 2 2 . 2 23 0.508

72 6 250 5.8 5.2-6.5 3.77 ±0.55 2.13 ±0.42 29.7 23 0.158

72 9 250 8.7 7.2 - 10.8 2.01 ±0.42 3.11 ±0.39 20.7 23 0.59896 6 250 5.6 5.0-6.3 3.52 ±0.53 2.36 ±0.41 27.4 23 0.240

96 9 250 8.5 7.1 - 10.3 2.15 ±0.42 3.01 ±0.39 2 0 . 8 23 0.594

Column headings are abbreviated as follows: h is the hours after treatment; n is the numbers of insects tested excluding the controls; 95 % CL is the 95 % confidence limits; Slope ± SE is the slope ± standard error; Intercept ± SE is the intercept ± standard error; y2 is the Pearson’s y? goodness-of-fit test; df is the degrees of freedom; P is the probability.

Table 4.8 were greater than 0.05 and so Pearson’s y? goodness-of-fit tests were not

significant, indicating that the data fit the assumptions of the probit model adequately.

At all mortality recording times, the LD5 0S estimated for female adults of CC were at

least 1 . 6 times greater than that estimated for the males, when doses expressed as

//g/insect. As at all mortality recording times the 95 % confidence limits did not

overlap it can be concluded that males were significantly more susceptible than

females of CC when doses were expressed as //g/insect. Moreover, the LD50s

estimated for female adults of CC were at least 1.4 times than those for the males,

when doses expressed as //g/mg of adult body weight. At 24, 72 and 96 h after

treatment, the 95 % confidence limits did not overlap, indicating that males were more

susceptible than females of CC at these mortality recording times. In contrast, at 48 h

109

after treatment the 95 % confidence limits overlapped, indicating that males and

females did not differ significantly in susceptibility at this time.

1 4.5.tsaeflu

4

3.5

3

- Males Females

1.5 2

Log (jig/insect)

2.5

>>6 ]

5.5 -

i 5 -0s 4.5

.tsJS 4 -2flu 3.5 -

3

1

B

MalesFemales

1.5 2

Log (jig/insect)

2.5

6 15.5 -

'H 5 -©E.ts

4.5 -

.0©h4 -

0. 3.5 -

3 -

MalesFemales

1.5 2

Log (jig/insect)

2.5

6 1 5.5 -

**5 -

Os 4.5 -.ts.0 4 -©ha. 3.5 -

3 -

D

MalesFemales

1.5 2

Log (jig/insect)

2.5

Figure 4.13 Log (//g/insect) - probit mortality lines calculated by probit analysis for

males and females of CC treated with the petroleum ether peel extract of CA using the


According to the results presented in Table 4.5, Table 4.6, Table 4.7 and Table 4.8, at

the end-point mortality of BO and CC (144 and 96 h after treatment, respectively)

males of both insect species were equally susceptible (either doses expressed as

//g/insect or //g/mg adult body weight) to the petroleum ether peel extract when

treated topically. In contrast, LD5 0S estimated for females of CC were about 2-fold

greater (doses expressed either as //g/insect or //g/mg adult body weight) than LD50S

estimated for females of BO. That results indicate that regardless of dose expression,

110

males of both species were equally susceptible (95 % confidence limits overlapped),

whereas BO females were more susceptible than CC females (95 % confidence limits

did not overlap) when treated topically with the petroleum ether peel extract of CA.

Moreover, CC adults reached the end-point mortality two days earlier than BO adults,

indicating that the extract acted more rapidly against CC adults.

4.5

- Males Females3.5

0 0.5 1 1.5

.ts©

6

5.5

5

4.5

4 -MalesFemales3.5

3

0 0.5 1 1.5

Log (jig/mg) Log (ng/mg)

I 4.5

I «uO* 3 5MalesFemales

0 0.5 1.51Log (*ig/mg)

.ts.o

6

5.5

5

4.5

4 - Males Females3.5

3

0 0.5 1 1.5

Log (ng/mg)

Figure 4.14 Log (wg/mg) - probit mortality lines calculated by probit analysis for

males and females of CC treated with the petroleum ether peel extract of CA using the


I l l

4.4 DISCUSSION

The results obtained by the two contact bioassays suggested that the petroleum ether

peel extract of CA is active to both fly species tested showing a clear

dose/concentration-dependent effect. However, in some cases, differences in

susceptibility were detected not only between the two species, but also between the

two sexes of the same species.

According to the results obtained by the Petri dish exposure bioassays males and

females of BO are equally susceptible. On the contrary, females were more tolerant

than males of CC. That difference in susceptibility between males and females of CC

(about 2-fold) could not be fully explained by the slightly greater body weight of

females. Moreover, personal observations during Petri dish exposure bioassays did

not reveal any difference in behaviour between males and females of CC. The most

plausible explanation for the higher tolerance of CC females is that females might

possess higher quantities of enzymes or more efficient mechanisms that can detoxify

the insecticidal active chemicals present in the petroleum ether peel extract. Weaver et

al. (1994) reported that males of the Mexican bean weevils were more susceptible

than females when exposed to filter papers treated with floral, foliar and root extracts

of Tagetes minuta. However, these authors attributed that difference to the smaller

size of males rather than to an innate difference in the mode of susceptibility.

Comparing the results obtained by the Petri dish exposure bioassays against BO and

CC adults, it is apparent that males and females of BO are significantly more

susceptible than males and females of CC, respectively. That difference in

susceptibility between the two species could not be fully explained by the slightly

112

greater body weight of males and females of CC compared to males and females of

BO, respectively. The most plausible explanation for the greater tolerance of males

and females of CC compared to males and females of BO, respectively is that CC

adults are less behaviourally active than BO adults and consequently they acquired

smaller quantities of the insecticidal active chemicals when the flies are placed in the

Petri dishes. That difference in the behavioural activity between BO and CC was

confirmed by personal observations during the Petri dish bioassays. Moreover, the

difference in the degree of susceptibility between the two sexes (males and females of

CC was 3.9 and 8.3 more tolerant than males and females of BO, respectively) could

be accounted to the higher quantities of enzymes or more efficient mechanisms of

females compared to males of CC. Abbassy et al. (1979) reported that Tribolium

confusum was significant more susceptible than Sitophilus granaries when lime,

lemon, sweet orange and native orange peel oils were tested using the Petri dish

exposure bioassay. However, when navel orange, grapefruit and mandarin peel oils

were tested the two insects were equally susceptible.

Bioassays performed in order to investigate whether mortality of BO and CC caused

in the Petri dish exposure bioassays was due to fumigation rather than contact, reveal

the absence of fumigant action. Moreover, even when the Petri dishes were sealed

with Parafilm, mortality of both species did not occur, indicating the relevant

unimportance of fumigant compared to contact activity of the petroleum ether peel

extract of CA. In contrast, Don-Pedro (1996) reported the relevant unimportance of

the contact toxicity of citrus oils. That discrepancy can be explained as Don-Pedro

used industrially extracted citrus peel oils, which contain mainly volatile compounds,

while volatiles of our extract were removed.

113

The results obtained from the topical application bioassays reveal that when doses

were expressed as //g/insect, males and females of BO are equally susceptible, which

is in agreement with the results obtained by the Petri dish exposure bioassay in which

no correction was done for the difference in body weight between males and females.

However, when doses were expressed as //g/mg adult body weight (corrected for the

difference in body weight between males and females), females are more susceptible

than males. Generally, female insects have a greater tolerance to insecticides, when

tests are corrected for any differences in body weight, which is attributable to their

metabolism (Matthews, 1997). That was not the case when BO was treated topically

with the petroleum ether peel extract. Since topical application bioassay is

independent of insect activity, the most possible explanation for the greater tolerance

of males could be that males possess higher quantities of enzymes or more efficient

mechanisms that can detoxify the insecticidal active chemicals present in the

petroleum ether peel extract.

Moreover, males of CC were more susceptible than females when treated topically

with the petroleum ether peel extract, either dose expressed as //g/insect or //g/mg

adult body weight. This finding is fully in accordance with the results obtained by the

Petri dish exposure bioassay, indicating that the explanation for the greater tolerance

of females compared to males of CC in both bioassays should be their ability to

metabolise the insecticidal active chemicals. Albrecht and Sherman (1987) found that

males of two Bactrocera species (B. dorsalis and B. curcub itae) were more

susceptible than females, whereas male and female of CC were equally susceptible,

when treated topically with avermectin Bi (doses expressed as //g/g). It is worth

mentioning that although male and female of CC were equally susceptible (their 95 %

114

confidence limits did not overlap), the LD50 estimated for males was 2-fold greater

than that estimated for females. It is obvious that generalisations about sex

susceptibility of insect species can not be made since different chemicals substances

produce different results although the same biological evaluation technique was used.

Comparing the results obtained by the topical application bioassays against BO and

CC adults it is apparent that males of both species are equally susceptible, whereas

females of BO are more susceptible than females of CC either dose expressed as

//g/insect or //g/mg adult body weight. The difference in the degree of susceptibility

between the two sexes (2-fold when doses expressed as //g/insect and 1.7-fold when

doses expressed as //g/mg adult body weight) when the two species were compared is

fully in accordance with the results of the Petri dish exposure bioassay. The higher

quantities of enzymes or more efficient mechanisms of females compared to males of

CC could explain not only the difference between BO and CC females, but also the

difference in the degree of susceptibility between the two sexes. Su et al. (1972b)

reported that although peel oils of lemon, grapefruit, lime, kumquat, tangerine,

orange, tangelo and temple orange were moderately toxic to rice weevil, none of these

oils showed any appreciable toxicity to black carpet beetle larvae, Indian meal moth

larvae, cigarette beetle, red flour beetle and confused flour beetle using the topical

application bioassay (doses expressed as //g/insect). Moreover, Albrecht and Sherman

(1987) reported great differences when the susceptibility of CC, B. dorsalis and B.

curcubitae of the same sex was compared; males of CC were 13.8 and 6.9 more

tolerant than B. dorsalis and B. curcubitae males, respectively while females of CC

were 48.3 and 23.2 more tolerant than B. dorsalis and B. curcubitae females,

respectively. Apart from the detoxification mechanisms, Stark and Sherman (1989)

115

suggested that significant differences in the ranking of toxicity between species can be

attributed to different penetration rates and excretion (Stark and Sherman, 1989).

Petri dish exposure and topical application bioassays produced different results when

the susceptibility of the same sex between the two fly species was compared.

Although in the topical application bioassay males of both species were equally

susceptible, in the Petri dish exposure bioassay BO males were 3.9 times more

susceptible than CC males. Moreover, females of BO were more susceptible than

females of CC in both bioassays, but the degree of susceptibility differed significantly

(8.3 and 2.0 times for the Petri dish exposure and topical application bioassay,

respectively). The greater activity of BO compared to the CC resulted in greater

susceptibility of BO compared to the CC in the Petri dish exposure bioassay, since

this bioassay is not independent of insect activity.

Slope values for the regression lines of both insect species were significantly greater

in the Petri dish exposure than in topical application bioassay. Since higher slopes

indicate a more sensitive technique (Dahm et al., 1961) it can be concluded that the

Petri dish exposure is more sensitive than the topical application bioassay when the

petroleum ether peel extract was tested against both BO and CC adults.

Petri dish exposure bioassay was not only more sensitive than topical application in

this study, but also in that bioassay the end-point mortality was reached earlier. In the

Petri dish exposure bioassay, BO and CC adults reached the end-point mortality two

and one days respectively earlier than in the topical application.

116

In both bioassays CC adults reached the end-point mortality earlier than BO adults. In

the Petri dish exposure bioassay CC adults reached the end-point mortality one day

earlier than BO adults, whereas in the topical application bioassay CC adults reached

the end-point mortality two days earlier than BO adults.

117

CHAPTER 5

ISOLATION OF INSECTICIDALLY ACTIVE COMPONENTS

FROM THE PETROLEUM ETHER PEEL EXTRACT OF CITRUS

AURANTIUM

5.1 INTRODUCTION

Fractionation is the process in which the components of a mixture, such as a plant

extract, can be separated into groups of compounds sharing similar physicochemical

characteristics (Houghton and Raman, 1998). Thus, solubility, volatility, polarity,

size, shape, electrical charge and functional groups present may influence the

grouping (Harwood et al., 1999). The methods used for achieving fractionation are:

precipitation, liquid-liquid partitioning, distillation, dialysis, chromatographic

techniques and electrophoresis. However, it should be noted that chromatography in

its various forms is the most common method and is exploited in most fractionation

procedures carried out nowadays (Houghton and Raman, 1998).

Chromatography depends upon the differential distribution of various components of

a mixture between two phases, the mobile phase and the stationary phase. The mobile

phase may be either a liquid or a gas and the stationary phase either a solid or liquid.

Various combinations of these components give the main types of chromatographic

techniques (Harwood et al., 1999). The main chromatographic techniques are: column

chromatography (gravity and flash), planar chromatography (thin layer and paper),

118

gas chromatography, high performance liquid chromatography (normal and reverse

phase), ion exchange chromatography and gel permeation chromatography.

Different fractionation strategies can be followed during a fractionation process.

Collecting large number of very small fractions means that its fraction is more likely

to contain a pure compound, but it requires more work for testing every fraction. This

also runs the risk of spreading the target compound over so many fractions that, if

originally present in only low concentrations, it may evade detection in any one of the

fractions. Collecting a few large, but relatively crude fractions resulted in more

fractionation steps, but the work in each step is considerably less (Cannell, 1998).

A common phenomenon during fractionation processes was discussed by Cannell

(1998) and Houghton and Raman (1998). They concluded that lack or disappearance

of activity occurred after fractionation may be due to the following reasons:

■ The active components may remain in the system used for fractionation.

■ The active components may have decomposed during the fractionation

process.

■ Observed activity in the crude extract is due to synergism between

components which are separated as a result of the fractionation process.

■ The active components may have been lost due to a process such as co

distillation or sublimation when liquid fractions are concentrated.

■ The starting sample was not prepared so as to be fully compatible with the

mobile phase, so that a large proportion of the active component precipitated

when loading on the top of the column.

119

■ Most of the active component(s) spread across a wide range of fractions in a

concentration too low to be detected by the assay.

Only a few efforts have been made to isolate the insecticidally active components of

promising Citrus extracts. Su (1976) reported the isolation of a single active

component from lyophilised hexane lemon peel oil. Fractionation was carried out by

three sequential TLC (silica gel F2 5 4) separations using three different developing

solvent systems. Fractionation was guided by topical application and the test insect

was adult cowpea weevil. However, although Su had achieved the isolation of a pure

active compound, she did not make any attempt to elucidate its structure. More

recently, Su and Horvat (1987) isolated and identified the four major components of

lyophilised hexane lemon peel oil. Fractionation was carried out by TLC (silica gel

F2 5 4) followed by HPLC (Cjg column) and was not guided by bioassays.

The majority of studies aiming to isolate the active ingredients from insecticidally

active plant extracts involve liquid-liquid partitioning, gravity column or flash column

chromatography, preparative thin layer chromatography and high performance liquid

chromatography. Dadang et al. (1996), using bioassay-guided procedures, isolated

from a methanol tuber extract of nutgrass an insecticidal sesquiterpenoid, which was

identified as a-cyperone (4, ll-selinadien-3-one). Fractionation was carried out by

using in sequence liquid-liquid partitioning (MeOH/n-hexane), gravity column

chromatography (silica gel), preparative TLC (silica gel F2 5 4) and finally HPLC.

Rodriguez-Saona et al. (1997, 1998) using flash chromatography on silica gel

separated an avocado idioblast cell oil into eight fractions. Two fractions were

revealed as promising against Spodoptera exigua. The most active fraction was

120

further purified by HPLC (silica column) resulting in the isolation and identification

of persin (Rodriguez-Saona et al., 1997). The other active fraction was further

fractionated by two sequential flash columns resulting in the isolation of several

insecticidal compounds (Rodriguez-Saona et al., 1998). Gbewonyo and Candy

(1992a) reported that gas chromatography allowed the purification of pellitorine from

a piper extract in a single step. However, they found that the activity of the extract

was decreased on passage through the GC column, which was attributed to the

instability of some insecticidal or synergistic compounds at the high temperatures

used. They concluded that GC is unsuitable for the separation and isolation of these

compounds. The same authors (1992b) reported the purification and identification of

five active amides of the same piper extract using different bioassay-guided

fractionation procedures. This time, fractionation was carried out by three sequential

TLC separations using three different developing solvent systems followed by

reversed-phase HPLC.

In this chapter the chromatographic procedure followed to isolate the bioactive

chemicals present in the petroleum ether peel extracts of CA and the obtained results

will be discussed.

121

5.2 METHODS

5.2.1 Fractionation procedure of the petroleum ether peel extract of CA

The active petroleum ether peel extract of CA obtained in section 4.2.1 was

fractionated using chromatographic techniques in order to isolate the insecticidally

active secondary metabolites it possessed. The fractionation procedure of the

petroleum ether peel extract of CA is shown in Figure 5.1 and was guided by Petri

dish exposure bioassays.

First gravity (open) column fractionation

Gravity column chromatography is the oldest form of chromatographic technique

(Houghton and Raman, 1998). Although it does not require special equipment and is

the most economic of the chromatographic techniques, as it requires only the most

basic grades of chromatographic adsorbents, it can give sufficient separation at the

initial fractionation steps (Harwood et al., 1999).

A glass column (500 mm x 45 mm i.d.) was used for the fractionation of 2 g of the

petroleum ether peel extract. This column was fitted at one end with a fine porosity

sintered glass disk above the stopcock in order to support the stationary phase. The

column was wet packed with 200 g of silica gel 60 for column chromatography (0.04-

0.063 mm/230-400 mesh ASTM, Macherey-Nagel, Germany). The 1 %

sample/stationary phase ratio used, offers sufficient separation while loss by

irreversible adsorption to the stationary phase was minimised. Prior to packing, silica

gel was suspended in DCM, which was the least polar solvent used for the

fractionation. The height of silica gel into the column was 32 cm. Wet packing was

used in preference to dry packing in order: (a) to achieve a narrow band of applied

122

Petroleum ether peel extract

First gravity column fractionation

F 2 F 3

Second gravity column fractionation

F 21 F 22 F 23 F 24

HPLC fractionation

F 221 F 222 F 223 F 224 F 225 F 226 F 227

Figure 5.1 Fractionation procedure of the petroleum ether peel extract of CA.

sample, (b) to achieve uniform flow of the mobile phase, and (c) to avoid air bubbles,

often trapped as the mobile phase passes through, which disrupt the flow (Houghton

and Raman, 1998).

As the force of the eluting solvents tends to disturb the flat surface of the stationary

phase, about 1 cm of anhydrous sodium sulphate was placed above the silica gel

surface in order to protect it. A petroleum ether peel extract solution (2 g/20 ml DCM)

was introduced on to the column. The stopcock was opened until the solution reached

the surface of the anhydrous sodium sulphate without touching it and then the

stopcock was closed. Above the sample, 1 cm of anhydrous sodium sulphate was

added in order to avoid sample dilution. Silica gel and anhydrous sodium sulphate

were washed with MeOH and then activated at 130 °C for 24 h prior to use. Three

fractions were collected by sequentially eluting the column with 1 1 each

(approximately 3 dead volumes) of DCM (Fi fraction), EtOAc (F2 fraction) and

MeOH (F3 fraction).

Second gravity column fractionation

The same column as above was used for the fractionation of 1 g of the active F2

fraction, obtained from the first gravity column fractionation. The column was wet

packed with 100 g silica gel (suspended in 20 % Hex in EtOAc). The height of silica

into the column was 16 cm. 1 g of the F2 fraction dissolved in 50 ml of 20 % Hex in

EtOAc was introduced on to the column. Four fractions were collected by sequentially

eluting the column with 500 ml each (approximately 3 dead volumes) of 20 % Hex in

EtOAc (F21 fraction), 40 % Hex in EtOAc (F22 fraction), 60 % Hex in EtOAc (F23

fraction) and 100 % MeOH (F24 fraction).

124

HPLC fractionation

High Performance Liquid Chromatography (HPLC) was used to fractionate further

the active F22 fraction, obtained from the second gravity column fractionation. HPLC

was used for the final stage of fractionation as it is a much more efficient technique

than gravity column chromatography and can lead to the isolation of pure substances.

There are many advantages of HPLC over conventional chromatography, and these

include increased speed, higher resolution, higher sensitivity, greater reproducibility,

and better sample recovery (Patel, 1997).

A binary LC pump 250 equipped with a LC-235 diode array detector (Perkin Elmer,

USA) was used. The Simple peak 202 (SRI Instruments, USA) software was

employed for recording the chromatograms. After testing, different columns,

stationary phases, mobile phases, elution profiles, flow rates, injection volumes and

UV detection wavelengths, the following chromatographic conditions were chosen for

performing the fractionation. A 250 mm x 10 mm i.d. LiChrospher 100 CN column

with a particle size of 10 //m (Merck, Germany) supplied with a 10 mm precolumn

packed with the same stationary phase was used. Two solvents, n-hexane (A) and

ethyl acetate (B), which were degassed by sonication before use, were selected. The

elution profile was: 0-35 min A:B 90:10 % (isocratic), 35-40 min from A:B 90:10 %

to 100 % B (linear gradient), 40-60 min 100 % B (isocratic). Prior to injection of

sample the column was re-equilibrated for 10 min at the initial conditions. The flow

rate used was 4 ml/min, the column temperature was ambient and the absorbance was

monitored at 310 nm (optical bandwidth 15 nm). The F22 fraction (1 mg/20 /d EtOAc

per injection) was introduced onto the column via a Rheodyne 7125 injector (USA)

fitted with a 20 ju\ sample loop. Seven fractions were collected (F22 1 , F22 2, F2 2 3, F2 2 4 ,

125

F2 2 5 , F226 and F22 7) (see Figure 5.4). Totally, 100 mg (100 runs) of the F22 fraction

were fractionated. All fractions were filtered, concentrated, weighed and stored in the

dark at -20 °C until used.

5.2.2 Bioassay of fractions

To monitor the activity of the fractions obtained during the fractionation procedure of

the petroleum ether peel extract of CA the Petri dish exposure bioassay was used (see

section 2.2.2). Five replicates of 10 (5 males and 5 females) 2-3 d-old BO flies each,

were employed for each treatment and the control. The 5 replicates were conducted on

5 successive days, using the same generation of flies. Different concentrations of

treatments were not run simultaneously, as screening experiments at lower

concentrations included only the more active treatments. Stock solutions were

prepared in DCM, as were subsequent dilutions. Fresh test solutions were prepared for

each replicate. Because the amount of the fractions obtained from the HPLC

fractionation was limited, during tests on these fractions, olive oil was added to each

treatment solution and control (DCM) so that each cm2 of the total inner surface of the

Petri dishes was layered with 0.1 /ri o f olive oil. Olive oil was used in order to reduce

the amount of the HPLC fractions tested, as it served as a carrier to facilitate pick up

of residues of the test materials from the Petri dish (Gbewonyo and Candy, 1992a).

The syringe used for the application of samples was thoroughly cleaned after the

application of each sample. Mortalities were recorded every 24 h for 3 d, when olive

oil was added to the treatments, whereas when olive oil was not added, mortalities

were recorded every 24 h for 4 d, after which there was no further increase in

mortality.

126

5.2.3 TLC analysis of fractions

TLC analyses were carried out as in section 3.2.3 in order to check the component

profile of the fractions obtained from the first gravity column fractionation against

that of the petroleum ether peel extract and moreover to check the component profile

of the fractions obtained from the second gravity column fractionation against that of

the F2 fraction. In each case fractions were checked using the same amount per spot

(100 ^g/spot for the first gravity column fractionation and 62.7 //g/spot for the second

gravity column fractionation) or equivalent amount to the initial extract/fraction

compared. The petroleum ether peel extract and all fractions were dissolved in DCM.

All analyses were performed using Hex:EtOAc (50:50, v/v) as the developing solvent

system. Photographs of the TLC plates were obtained by the Camedia digital camera

under 254 nm UV light

5.2.4 HPLC analysis of fractions

HPLC analyses were carried out in order to investigate whether the three major peaks

of the F22 fraction were separated adequately during the HPLC fractionation. All

chromatographic conditions were the same as in section 5.2.1 except the elution

profile which was Hex:EtOAc (90:10, v/v) for all the time required for the analyses of

fractions F2 2 1 , F22 2, F22 3, F22 4 , F225 and F2 2 6- However, when analysing the F227 fraction

the same elution profile as in section 5.2.1 was set. Each fraction was analysed in

triplicate. Fraction solutions analysed were equivalent to 200 //g/injection of the F22

fraction. All fractions were dissolved in EtOAc. Percentages of the three major peaks

present in each fraction obtained from the FIPLC fractionation were calculated.

127

5.3 RESULTS

5.3.1 F irst gravity column fractionation

Three fractions (Fi, F2 and F3) were obtained from the first gravity column

fractionation of the petroleum ether peel extract of CA (see section 5.2.1) in order to

detect whether the insecticidally active chemicals present in the petroleum ether peel

extract were recovered during the first fractionation step in a single fraction.

Recovery

The weights of fractions and their % recovery after the first gravity column

fractionation of the petroleum ether peel extract of CA, are shown in Table 5.1. The

amount recovered in the F2 fraction was 2.3 and 9.1 times greater than that of the Fi

and F3 fractions, respectively. The low amount recovered in the F3 fraction can be

explained, by the use of a non-polar solvent such as petroleum ether for the extraction

of peels. Chemicals extracted from peels of CA using petroleum ether were mainly

non-polar, unionised substances, which were already eluted from the column

(fractions Fi and F2) prior elution with MeOH. It must be noticed that 3.3 % ( 6 6 mg)

of the petroleum ether peel extract were not eluted from the column. These chemicals

were mainly the more polar chemicals present in the petroleum ether peel extract and

Table 5.1 Weights of fractions and their % recovery after the first gravity column

fractionation of 2 g of the petroleum ether peel extract of CA.

Fraction Elutionsolvent

Amount recovered(mg)

Recovery(%)

Fi DCM 542 27.1

f2 EtOAc 1254 62.7

f3 MeOH 138 6.9Sum of fractions 1934 96.7

128

were adsorbed irreversibly to the stationary phase.

TLC analysis

The component profile of Fj, F2 , F3 fractions was checked against that of the

petroleum ether peel extract by TLC using Hex:EtOAc (50:50, v/v) as the developing

solvent system. In Figure 5.2A, 100 fig and the petroleum ether peel extract and 27.1,

62.7, 6.9 fig (equivalent to 100 fig of the petroleum ether peel extract) of the Fi, F2 , F3

fractions, respectively were spotted on the baseline of the TLC plate. In Figure 5.2B,

100 fig of the Fi, F2, F3 fractions and the petroleum ether peel extract were spotted.

Figures 5.2A and 5.2B indicate that the first gravity column effectively fractionated

the petroleum ether peel extract as different types of compounds based on their

polarity were present in each different fraction. Moreover, Figure 5.2A indicates that

decomposition during fractionation had not occurred, as compounds present in the

petroleum ether peel extract had not disappear, and moreover, compounds not present

in the petroleum ether peel extract did not appear, after fractionation.


The petroleum ether peel extract and its fractions obtained from the first gravity

column fractionation were tested for activity using the Petri dish exposure bioassay in

order to detect which fraction contain the insecticidal active chemicals. The initial

screening concentration was set at 100 //g/cm2. Only fractions that exhibited activity

at this concentration were further tested at 50 //g/cm2.

129

1.00 —

0.95 -

0.90 —

0.85 -

0.80 —

0.75 -

0.70 —

0.65 -

0.60 —

0.55 -

0.50 —

0.45 -

0.40 —

0.35 -

0.30 —0.25 -

0.20 —

0.15 -

0.10 —

0.05 -

0.00 —

1.00 —

0.95 -

0.90 —

0.85 -

0.80 —

0.75 -

0.70 —

0.65 -

0.60

0.55 -

0.50 —

0.45 -

0.40 —

0.35 -

0.30 —

0 .25-

0.20 —

0.15-

0.10 —

0.05 -

0.00 —

PE F, F2 F3

Figure 5.2 Thin layer chromatograms of the fractions obtained from the first gravity

column fractionation of the petroleum ether peel extract.

PE: Petroleum ether peel extract; Fi, F2 and F3: DCM, EtOAc and MeOH fractions, respectively; R f: Retention factor; (A) PE: 100 //g/spot; Fj, F2 and F3: 27.1, 62.7 and 6.9 //g/spot, respectively (equivalent to 100 //g/spot o f PE); (B) PE, Fj, F2 and F3: 100 /ig/spot. Developing solvent system: Hex:EtOAc (50:50, v/v). UV light: 254 nm.

The results of the two experiments are presented in Table 5.2. The only active fraction

at 100 /ig/cm was the F2 fraction, whereas the Fi and F3 fractions were inactive.

Moreover, the activity of the petroleum ether peel extract was significantly greater

than that of the F2 fraction at both concentrations tested at all mortality recording

times.

Results indicate that the activity of the petroleum ether peel extract cannot be fully

explained by the activity of the F2 fraction. There are three possible explanations that

130

can be proposed for the reduced activity of the F2 fraction compared to the activity of

the petroleum ether peel extract: (a) one or more active components were irreversible

adsorbed in the stationary phase and consequently were not eluted from the column,

(b) one or more active components decomposed during fractionation, and (c) observed

activity in the petroleum ether peel extract was due to synergism among components

which were separated in different fractions as a result of the fractionation process.


extract of CA and its fractions obtained from the first gravity column fractionation


Concentration//g/cm2 Fraction

% MortalityMean8 ± SEb, at hours after treatment

24 48 72 96PEC 46 ± 12.1a 98 ± 2.0a 100 ± 0.0a 100 ± 0.0a

Fi 0 ± 0.0b 0 ± 0.0c 0 ± 0.0c 0 ± 0.0c100 f2 8 ± 5.8b 46 ± 6.8b 60 ± 8.9b 64 ± 8.7b

f3 2 ± 2.0b 2 ± 2.0c 2 ± 2.0c 2 ± 2.0cControl*1 2 ± 2.0b 2 ± 2.0c 2 ± 2.0c 4 ± 2.4c

PE 30 ± 5.5a 74 ± 6.0a 96 ± 2.4a 98 ± 2.0a50 f2 2 ± 2.0b 10 ± 3.2b 12 ± 3.7b 14 ± 5.1b

Control 0 ± 0.0b 2 ± 2.0b 2 ± 2.0b 2 ± 2.0b

“Mean o f 5 replicates o f 10 insects each (5 males and 5 females). bSE: Standard error o f mean. 'PE: Petroleum ether peel extract. dTreated with DCM. The two experiments (100 and 50 //g/cm2) were not run simultaneously. Means within a column o f a single concentration followed by the same letter are not significantly different (P - 0.05, Tukey’s HSD test). Mortalities were transformed before ANOVA to arcsine V p, where p is the proportion o f dead insects. Means ± SE o f untransformed data are reported.

In order to clarity the reason for the reduced activity of the F2 fraction, an experiment

was designed in which the activity of the petroleum ether peel extract was compared

to the activity of the sum of the three fractions. The contribution of each fraction to

the concentrations of the sum of fractions tested was equivalent to the screening

131

concentration of the petroleum ether peel extract. In detail, at the screening

concentration of 50, 25 and 15 //g/cm2 the concentration of the Fi fraction was 13.6,

2 2 6 . 8 and 4.1 //g/cm , the concentration of the F2 fraction was 31.4, 15.7 and 9.4 //g/cm

while the concentration of the F3 fraction was 3.4,1.7 and 1.0 //g/cm2, respectively.

The results of the three experiments are shown in Table 5.3. There were no significant

differences in activity between the petroleum ether peel extract of CA and the sum of

fractions (Fi + F2 + F3) at the three concentrations tested at all mortality recording

times.

The results indicate that active chemicals were not decomposed or remain irreversibly

adsorbed to the stationary phase during the fractionation process. Moreover, it was

evident that synergism between chemicals of different fractions took place, which

explains why the activity of the F2 fraction was significantly lower than that of the

petroleum ether peel extract.

A further experiment was designed in order to detect whether the activity of the

petroleum ether peel extract could be fully explained by the activity of a binary blend

of the three fractions. The petroleum ether peel extract and the three binary blends of

fractions (Fi + F2, Fi + F3 and F2 + F3) were tested. The contribution of each fraction

to the concentration of each binary blend was equivalent to the screening

concentration of the petroleum ether peel extract (for details see previous experiment).

Only binary blends that were active at the initial experiment were further tested.

132


extract of CA and the sum of its fractions obtained from the first gravity column

fractionation using the Petri dish exposure bioassay.

Concentrationpg/cm2

% MortalityTreatment Mean8 ± SEb, at hours after treatment

24 48 72 96

1st Experiment50 PEC 34 ± 9.3a 76 ± 7.5a 98 ± 2.0a 98 ± 2.0a

13.6 + 31.4 + 3.4 F!+F2+F3 32 ± 9.2a 82 ± 3.7a 100 + 0.0a 100 ± 0.0a0 Controld 2 ± 2.0b 4 ± 2.4b 4 ± 2.4b 4 ± 2.4b

2nd Experiment25 PE 6 + 4.0a 58 + 7.3a 74 ± 7.5a 78 + 5.8a

6.8+15.7+1.7 F 1+F2 +F3 10 + 4.5a 50 + 7.1a 68 ± 5.8a 70 ± 5.5a0 Control 0 ± 0.0a 2 ± 2.0b 2 ± 2.0b 2 ± 2.0b

3 rd Experiment15 PE 2 ± 2.0a 16 + 4.0a 28 ± 6.6a 32 ± 8.0a

4.1 +9.4+ 1.0 F 1+F2 +F 3 4 ± 2.4a 18 + 3.7a 36 + 5.la 38 ± 5.8a0 Control 2 ± 2.0a 2 ± 2.0b 4 ± 2.4b 4 ± 2.4b

“Mean o f 5 replicates o f 10 insects each (5 males and 5 females). bSE: Standard error o f mean. °PE: Petroleum ether peel extract o f CA. dTreated with DCM. The three experiments were not run simultaneously. Means within a column o f a single experiment followed by the same letter are not significantly different (P = 0.05, Tukey’s HSD test). Mortalities were transformed before ANOVA to arcsine V/?, where p is the proportion o f dead insects. Means ± SE o f untransformed data are reported.

The results of the two experiments are displayed in Table 5.4. The Fi + F3 and F2 + F3

binary blends were inactive at 25 pg/cm2, whereas the petroleum ether peel extract

and the Fi + F2 binary blend were active and did not differ significantly at both

experiments at all mortality recording times.

The results of these experiments indicate that synergism occurred between chemicals

that were eluted during fractionation to the Fi and F2 fractions. The activity of the

petroleum ether peel extract of CA can be fully ascribed to the activity of the Fi + F2

133

binary blend. Moreover, the chemicals present in the F3 fraction did not have any

effect to the insecticidal activity of the petroleum ether peel extract.


extract of CA and different binary blends of its fractions obtained from the first

gravity column fractionation using the Petri dish exposure bioassay.

Concentration/zg/cm2 Treatment

% MortalityMean*1 ± SEb, at hours after treatment

24 48 72 961st Experiment

25 PEC 10 ± 5.5ab 48 + 7.3a 62 ± 9.7a 64 ± 8.7a6.8+15.7 Fi+F2 10 + 3.2a 38 ± 9.2a 56 + 8.1a 60 + 7.1a6.8+1.7 F 1+F3 2 ± 2.0ab 2 ± 2.0b 2 ± 2.0b 2 ± 2.0b

15.7+1.7 F2 +F 3 0 ± 0.0b 2 ± 2.0b 2 ± 2.0b 2 ± 2.0b0 Control 0 ± 0.0b 0 ± 0.0b 2 ± 2.0b 2 ± 2.0b

2nd Experiment15 PE 6 ± 4.0a 14 ± 5. lab 26 ± 5.1a 30 ± 3.2a

4.1 +9.4 Fi+F2 2 ± 2.0a 20 + 4.5a 34 + 5.1a 36 ± 6.0a0 Control 2 ± 2.0a 2 ± 2.0b 2 ± 2.0b 2 ± 2.0b

“Mean o f 5 replicates o f 10 insects each (5 males and 5 females). bSE: Standard error o f mean. CPE: Petroleum ether peel extract o f CA. dTreated with DCM. The two experiments were not run simultaneously. Means within a column o f a single experiment followed by the same letter are not significantly different ( P = 0.05, Tukey’s HSD test). Mortalities were transformed before ANOVA to arcsine Vp, where p is the proportion o f dead insects. Means ± SE o f untransformed data are reported.

5.3.2 Second gravity column fractionation

Bioassays conducted with the fractions obtained from the first gravity column indicate

that the only fraction which exhibit activity when tested individually was the F2

fraction. Hence, fractionation was continued with this fraction. Four fractions (F2 1 ,

F2 2, F2 3 and F2 4) were obtained from the second gravity column fractionation of the F2

134

fraction (see section 5.2.1) in order to detect the fraction that contain the insecticidal

active chemicals present in the F2 fraction.

Recovery

The weights of fractions and their % recovery after the second gravity column

fractionation of the F2 fraction are shown in Table 5.5. The highest amount of

chemicals of the F2 fraction was recovered in the F23 fraction (633 mg). The amount

of chemicals recovered in fraction F22 was 2 . 6 times less than that recovered in the F23

fraction, but 3.0 and 5.9 times more than that recovered in the F24 and F21 fractions,

respectively. 99.7 % of the total amount of the F2 fraction placed into the column was

recovered in the four fractions (F21 , F22 , F23 and F2 4) collected. The 0.3 % (3 mg) of

the F2 fraction that seems not to be present in the fractions could be attributed to

uncertainties associated with weighing rather than irreversible adsorption to the

stationary phase.

Table 5.5 Weights of fractions and their % recovery after the second gravity column

fractionation of 1 g of the F2 fraction.

Fraction Elution solvents Amount recovered (mg)

Recovery(%)

F21 Hex:EtOAc (80:20, v/v) 41 4.1

F22 Hex:EtOAc (60:40, v/v) 243 24.3

F23 Hex.EtOAc (40:60, v/v) 633 63.3

F24 MeOH 80 8.0Sum of fractions 997 99.7

135

TLC analysis

The component profile of F2 1 , F2 2 , F23 and F24 fractions was checked against that of

the F2 fraction by TLC using Hex:EtOAc (50:50, v/v) as the developing solvent

system. In Figure 5.3A, 62.7 //g of the F2 and 2.6, 15.2, 39.7, 5.0 //g (equivalent to

62.7 fig of the F2 fraction) of the F2 1 , F2 2 , F23 and F24 fractions, respectively were

spotted on the baseline of the TLC plate. In Figure 5.3B, 62.7 fig of the F2 , F21 , F22 ,

F23, F24 fractions were spotted. Figures 5.3A and 5.3B indicate that the second gravity

Rr1.00 — A 1.00 —

0.95 - 0.95 -

0.90 — 0.90 —0.85 - 0.85 -

0.80 — 0.80 —0.75 - 0.75 -

0.70 — 0.70 —0.65 - 0.65 -

0.60 — 0.60 —0.55 - 0.55 -

0.50 — 0.50 —0.45 - 0.45 -

0.40 — A A 0.40 —0.35 - 0.35 -

0.30 — 0.30 —025 - 025 -

0.20 — 0.20 —0.15 - 0 .15-

0.10 — 0.10 —0.05 - 0.05 -

0.00 — 0.00 —

21 F 22 *23 24

Figure 5.3 Thin layer chromatograms of the fractions obtained from the second

gravity column fractionation of the F2 fraction.

F2: EtOAc fraction; F2!, F22, F^ and F^: 20 % EtOAc, 40 % EtOAc, 60 % EtOAc and MeOH fractions, respectively. R f: Retention factor. (A) F2: 62.7 //g/spot; F2 i, F22 F23 and F2 4: 2.6, 15.2, 39.7 and 5.0 //g/spot, respectively (equivalent to 62.7 //g/spot o f F2). (B) F2, F2 j, F22, F2 3 and F24: 62.7 //g/spot. Developing solvent system: Hex:EtOAc (50:50, v/v). UV light: 254 nm.

136

column provided sufficient fractionation, since different types of compounds based on

their polarity were present in each individual fraction. Moreover, Figure 5.3A

indicates that decomposition during fractionation had not occurred, as compounds

present in the F2 fraction did not disappear, and moreover, compounds not present in

the F2 fraction did not appear, after fractionation.


The F2 fraction and its fractions obtained from the second gravity column

fractionation of the F2 fraction were tested for activity using the Petri dish exposure

bioassay in order to detect in which fraction the active chemicals were eluted. The

initial screening concentration was set at 100 //g/cm2. Only fractions that exhibited

activity at this concentration were further tested at 50 /^g/cm2.

The results of the two experiments are presented in Table 5.6. The activity of the F22

fraction was significantly greater than the activity of the F2 fraction at both

concentrations tested at all mortality recording times. Moreover, fractions F2 1 , F23 and

F24 were inactive even at 100 /*g/cm2.

That results indicate that the insecticidally active chemicals of the F2 fraction were

eluted to the F22 fraction during the second fractionation step. However, in order to

investigate whether the activity of the F2 fraction could be fully explained by the

activity of the F22 fraction, a further experiment was designed. In this experiment the

insecticidal activity of the F2 fraction at 100 and 75 /*g/cm2 was compared to the

activity of the equivalent concentration of the F22 fraction (24.30 and 12.15 //g/cm2),

respectively.

137

Table 5.6 Percentage mortality of BO adults exposed to the F2 fraction and its

fractions obtained from the second gravity column fractionation using the Petri dish

exposure bioassay.

Concentration/^g/cm2

% MortalityFraction Meana ± SEb, at hours after treatment

24 48 72 96

f2 12 ± 3.7b 30 ± 4.5b 56 ± 5.1b 58 ± 6 .6 b

F2i 2 ± 2 .0 b 2 ± 2 .0 c 4 ± 2.4c 4 ± 2.4c

F22 74 ± 8.1a 94 ± 2.4a 98 ± 2.0a 1 0 0 ± 0 .0 a1 0 0

F23 0 ± 0 .0 b 0 ± 0 .0 c 0 ± 0 .0 c 0 ± 0 .0 c

F24 2 ± 2 .0 b 2 ± 2 .0 c 2 ± 2 .0 c 2 ± 2 .0 c

Control0 0 ± 0 .0 b 2 ± 2 .0 c 2 ± 2 .0 c 2 ± 2 .0 c

f2 2 ± 2 .0 b 4 ± 2.4b 6 ± 4.0b 8 ± 3.7b50 F22 32 ± 6 .6 a 6 8 ± 5.8a 96 ± 2.4a 1 0 0 ± 0 .0 a

Control 0 ± 0 .0 b 0 ± 0 .0 b 0 ± 0 .0 b 0 ± 0 .0 b

“Mean o f 5 replicates o f 10 insects each (5 males and 5 females). bSE: Standard error o f mean. 'Treated with DCM. The two experiments (100 and 50 //g/cm2) were not run simultaneously. Means within a column o f a single concentration followed by the same letter are not significantly different (P = 0.05, Tukey’s HSD test). Mortalities were transformed before ANOVA to arcsine V/?, where p is the proportion o f dead insects. Means ± SE o f untransformed data are reported.

The results of the two experiments are presented in Table 5.7. The activity of the F22

fraction did not differ significantly to the activity of the F2 fraction at both

experiments at all mortality recording times, indicating that only the active chemicals

present in the F22 fraction were responsible for the observed activity of the F2 fraction.

Chemicals present in the F2 1, F23 and F24 fractions made no contribution to the

insecticidal activity of the F2 fraction.

138

Table 5.7 Percentage mortality of BO adults exposed to the F2 and F22 fractions using


Concentrationpg/cm2

% MortalityFraction Mean® ± SEb, at hours after treatment

24 48 72 96

1 0 0 . 0 0 f2

1st Experiment

12 ± 3.7a 44 ± 6 .8 a 56 ± 6.0a 58 ± 4.9a

24.30 F22 10 ± 3.2a 44 ± 7.5a 64 ± 8.1a 6 6 ± 6 .8 a

0 . 0 0 Control0 2 ± 2 .0 a 2 ± 2 .0 b 2 ± 2 .0 b 2 ± 2 .0 b

50.00 f2

2nd Experiment

4 ± 2.4a 20 ± 3.2a 30 ± 5.5a 32 ± 5.8a12.15 F22 6 ± 4.0a 26 ± 5.1a 30 ± 5.5a 30 ± 5.5a

0.00 Control 0 ± 0 .0 a 0 ± 0 .0 b 2 ± 2 .0 b 2 ± 2 .0 b

“Mean o f 5 replicates o f 10 insects each (5 males and 5 females). bSE: Standard error o f mean. 'Treated with DCM. The two experiments were not run simultaneously. Means within a column o f a single experiment followed by the same letter are not significantly different (P = 0.05, Tukey’s HSD test). Mortalities were transformed before ANOVA to arcsine V/?, where p is the proportion o f dead insects. Means ± SE o f untransformed data are reported.

5.3.3 HPLC fractionation

Seven fractions (F2 2 1 , F2 2 2, F223 F2 2 4, F2 2 5 , F226 and F2 2 7) were obtained from the HPLC

fractionation of the F22 fraction as shown in Figure 5.4 in order to isolate the

compound(s) responsible for the activity of the F22 fraction.

Recovery

The weights of fractions and their % recovery after the HPLC fractionation of the F22

fraction are shown in Table 5.8. There was no fraction that accounted for less than 9.1

% and more than 22.3 % of the amount of the F22 fraction fractionated. The amount

recovered in the F226 fraction (highest recovery) was 2.5 times greater than the amount

recovered in the F222 and F224 fractions (lower recovery). 99.8% of the total amount

139

Retention time

Figure 5.4 HPLC chromatogram of the F22 fraction.

Column: 250 x 10 mm i d. supplied with a 10 mm precolumn; stationary phase: LiChrospher 100 CN with a particle size of 10 ftm\ mobile phase: n-hexane (A) and ethyl acetate (B); elution profile: 0-35 min A:B 90:10 % (isocratic), 35-40 min from A:B 90:10 % to 100 % B (linear gradient), 40-60 min 1 0 0 % B (isocratic); flow rate: 4 ml/min; column temperature: ambient; UV detection at 310 nm; amount injected: 1 mg/20//l EtOAc; 1, 2, 3, 4, 5, 6 and 7: F221, F222, F223, F224, F225, F226 and F227 fractions, respectively.

of the F22 fraction placed into the column was recovered in the HPLC fractions

collected. The 0.2 % (2 mg) of the F22 fraction that seems not to be present in the

fractions could be attributed to uncertainties associated with weighing rather than

irreversible adsorption to the stationary phase.

Table 5.8 Weights of fractions and their % recovery after the HPLC fractionation of

100 mg of the F22 fraction.

Fraction Retention time (min) Amount recovered (mg) Recovery (%)

F221 0.00 -12.7 18.9 18.9

F222 12.7 - 14.0 9.1 9.1

F223 14.0-21.0 18.0 18.0

F224 21.0-22.8 9.1 9.1

F225 22.8 -31.5 9.4 9.4

F226 31.5-34.3 22.3 22.3

F227 34.3 - 60.0 13.0 13.0Sum of fractions 99.8 99.8

HPLC analysis

F2 2 1 , F22 2, F223 F22 4 , F2 2 5, F226 and F227 fractions were analysed by HPLC in order to

investigate whether the three major peaks of the F22 fraction were separated

adequately during the HPLC fractionation. The results obtained from the analyses

(Table 5.9) reveal that the three major peaks were separated adequately and recovered

in more than 94.2 % in a single fraction indicating that loss of these peaks

(distribution to other fractions) during their isolation was kept to minimum.

141

Table 5.9 Distribution (%) of the three major peaks of fraction F22 to the fractions

obtained from the HPLC fractionation of 100 mg of the F22 fraction.

FractionMajor peaks retention time (min)

13.2 21.7 32.5

% Distribution (mean8 ± SEb)

F221 1.7 ±0.07 - -

F222 96.8 ±0.21 - -

F223 1.5 ± 0.17 2.8 ±0.15 -

F224 - 94.2 ±0.47 -

F225 - 3.1 ±0.35 2.3 ±0.12

F226 - - 95.3 ±0.21

F227 - - 2.4 ±0.10

“Mean of 3 replicates. bSE: Standard error of mean.


The F22 fraction and its fractions obtained from the HPLC fractionation were tested

for activity using the Petri dish exposure bioassay in order to isolate the active

compound(s) present in the F22 fraction. The initial screening concentration was set at

5 //g/cm2. Only fractions that exhibited activity at this concentration were further

tested at 2.5 //g/cm2. At both experiments olive oil was added (0.65 % v/v) to each

fraction solution and the control.

The results of the two experiments are presented in Table 5.10. The only active

fractions at 5 //g/cm2 were the initial F22 and the recovered from the HPLC F2 2 6 ,

whereas all the other HPLC fractions were inactive. Moreover, the activity of the F226

fraction was significantly greater than that of the F22 at the concentration of 2.5

//g/cm2 at all mortality recording times.

142

Table 5.10 Percentage mortality of BO adults exposed to the F22 fraction and its

fractions obtained from the HPLC fractionation using the Petri dish exposure

bioassay.


% MortalityFraction Mean8 ± SEb, at hours after treatment

24 48 72

F22 60 ± 5.5b 94 ± 2.4a 98 ± 2.0a

F221 2 ± 2.0c 2 ± 2.0b 4 ± 2.4b

F222 0 ± 0.0c 0 ± 0.0b 0 ± 0.0b

F223 0 ± 0.0c 0 ± 0.0b 0 ± 0.0b5 F224 2 ± 2.0c 2 ± 2.0b 2 ± 2.0b

F225 0 ± 0.0c 0 ± 0.0b 0 ± 0.0b

F226 82 ± 5.8a 96 ± 2.4a 100 ± 0.0a

F227 2 ± 2.0c 2 ± 2.0b 2 ± 2.0b

Control0 0 ± 0.0c 2 ± 2.0b 2 ± 2.0b

F22 14 ± 5.1b 20 ± 6.3b 20 ± 6.3b2.5 F226 36 ± 6.8a 64 ± 7.5a 68 ± 5.8a

Control 0 ± 0.0c 0 ± 0.0c 0 ± 0.0c

*Mean of 5 replicates of 10 insects each (5 males and 5 females). bSE: Standard error of mean. 'Treated with 1 ml of DCM plus olive oil (0.65 % v/v). The two experiments (5 and 2.5 //g/cm2) were not run simultaneously. Means within a column of a single concentration followed by the same letter are not significantly different (P = 0.05, Tukey’s HSD test). Mortalities were transformed before ANOVA to arcsine where p is the proportion of dead insects. Means ± SE of untransformed data are reported.

That results indicate that the insecticidally active chemicals of the F22 fraction were

eluted to the F226 fraction during the HPLC fractionation. However, in order to

investigate whether the activity of the F22 fraction could be fully explained by the

activity of the F226 fraction a further experiment was designed. In that experiment

(Table 5.11) the insecticidal activity of the F22 fraction at 5.0 /*g/cm2 was compared to

the activity of the equivalent concentration of the F226 fraction (1.1 y«g/cm ). F22

fraction was significantly more active than the F226 fraction at all mortality recording

times indicating that the activity of the F22 fraction cannot be fully explained by the

activity of the F226 fraction. Two explanations can be proposed for the reduced

143

activity of the F226 fraction compared to the activity of the F22 fraction: (a) one or

more active components were decomposed during fractionation, and (b) observed

activity of the F22 fraction is due to synergism among components which were

separated in different fractions as a result of the fractionation process.

Table 5.11 Percentage mortality of BO adults exposed to the F22 and F22 6 fractions



% MortalityFraction Mean8 ± SEb, at hours after treatment

24 48 725.0 F22 72 ± 7.3a 94 ± 4.0a 96 ± 2.40a1.1 F226 2 ± 2.0b 4 ± 2.4b 6 ± 2.4b0.0 Control0 2 ± 2.0b 2 ± 2.0b 2 ± 2.0b

“Mean of 5 replicates of 10 insects each (5 males and 5 females). bSE: Standard error of mean. cTreated with 1 ml of DCM plus olive oil (0.65 % v/v). Means within a column followed by the same letter are not significantly different (P = 0.05, Tukey’s HSD test). Mortalities were transformed before ANOVA to arcsine Vp, where p is the proportion of dead insects. Means ± SE of untransformed data are reported.

In order to clarify the reason for the reduced activity of the F226 fraction a set of

experiment was conducted in order to compare the activity of the F22 fraction against

that of the sum of its fractions obtained from the HPLC fractionation. The F22 fraction

and the sum of its fractions were tested at three concentrations (5, 4 and 3 //g/cm2).

The contribution of each fraction to the concentrations of the sum of fractions tested

was equivalent to the screening concentration of the F22 fraction. In detail, at the

screening concentration of 5, 4 and 3 //g/cm2, the concentration of the F221 fraction

was 0.945, 0.756 and 0.567 //g/cm2, the concentration of the F222 fraction was 0.455,

0.364 and 0.273 //g/cm2, the concentration of the F223 fraction was 0.900, 0.720 and

0.540 //g/cm2, the concentration of the F224 fraction was 0.455, 0.364 and 0.273

//g/cm2, the concentration of the F225 fraction was 0.470, 0.376 and 0.282 //g/cm2, the

144

concentration of the F226 fraction was 1.115, 0.892 and 0.669 //g/cm , and the

concentration of the F227 fraction was 0.650, 0.520 and 0.390 //g/cm2, respectively.

The results of the three experiments are shown in Table 5.12. There were no

significant differences in activity between the F22 fraction and the sum of its fractions

(F221 + F222 + F223 + F224 + F225 + F226 + F2 2 7) at the three concentrations tested at all

mortality recording times.

The results clearly show that active chemicals were not decomposed during the HPLC

fractionation. Moreover, it was evident that synergism between chemicals present in

different fractions took place which can fully explained why the activity of the F226

fraction was significant lower than that of the F22 fraction.

A further set of experiments was conducted in order to detect among which fractions

synergism occurred. The F22 fraction and three different combinations of the HPLC

fractions were tested. These combinations were: the active F22 6 together with the two

other major peaks (F222 + F224 + F22 6); the active F226 together with all other fractions

except the two other major peaks (F221 + F223 + F225 + F2 26 + F2 2 7); and all fractions

except the active F226 (F221 + F222 + F223 + F224 + F225 + F2 2 7). The contribution of each

fraction to the concentration of each combination was equivalent to the screening

concentration of the F22 fraction, 4 and 3 //g/cm2, (for details see previous

experiment). Only fractions combinations that were active in the initial experiment

were tested further.

145

Table 5.12 Percentage mortality of BO adults exposed to the F22 fraction and the sum of fractions obtained from the HPLC fractionation using



Treatment

% Mortality

Mean® ± SEb, at hours after treatment

24 48 72

1st Experiment5.000 F22 62 ± 6 .6 a 8 8 ± 3.7a 90 ± 3.2a

0.945 + 0.455 + 0.900 + 0.455 + 0.470 + 1.115 + 0.650 F22I + F222 + F223 + F224 + F225 + F226 + F ^ 64 ± 5.1a 92 ± 2.0a 96 ± 2.4a0 . 0 0 0 Control' 0 ± 0 .0 b 0 ± 0 .0 b 0 ± 0 .0 b

2nd Experiment

4.000 F22 56 ± 7.5a 80 ± 3.2a 80 ± 3.2a0.756 + 0.364 + 0.720 + 0.364 + 0.376 + 0.892 + 0.520 F221 + F222 + F223 + F224 + F225 + F226 + F227 50 ± 4.5a 72 ± 9.7a 76 ± 7.5a

0 . 0 0 0 Control 0 ± 0 .0 b 2 ± 2 .0 b 2 ± 2 .0 b3 rd Experiment

3.000 F22 18 ± 5.8a 36 ± 4.0a 38 ± 4.9a

0.567 + 0.273 + 0.540 + 0.273 + 0.282 + 0.669 + 0.390 F22I + F222 + F223 + F224 + F225 + F226 + F227 20 ± 3.7a 36 ± 6 .8 a 36 i 6 .8 a

0 . 0 0 0 Control 2 ± 2 .0 b 2 ± 2 .0 b 2 ± 2 .0 b

“Mean o f 5 replicates o f 10 insects each (5 males and 5 females). bSE: Standard error o f mean. 'Treated with 1 ml o f DCM plus olive oil (0.65 % v/v). The three experiments were not run simultaneously. Means within a column o f a single experiment followed by the same letter are not significantly different (P = 0.05, Tukey’s HSD test). Mortalities were transformed before ANOVA to arcsine 'Ip, where p is the proportion o f dead insects. Means ± SE o f untransformed data are reported.

The results of the two experiments are displayed in Table 5.13. The F221 + F223 + F225

+ F226 + F227 and F221 + F222 + F223 + F224 + F225 + F227 combinations of fractions were

inactive at the initial experiment, whereas the F22 fraction and the F222 + F224 + F226

combination of fractions were active and did not differ significantly at both

experiments at all mortality recording times.


that were eluted during the HPLC fractionation to the F2 2 2 , F224 and F226 fractions. The

activity of the F22 fraction can be fully ascribed to the activity of the F222 + F224 + F226

blend. Moreover, the chemicals present in the F22 1 , F22 3 , F225 and F227 fractions did not

have any contribution to the insecticidal activity of the F22 fraction.

A further set of experiments was conducted in order to detect whether the activity of

F22 fraction could be fully explained by the activity of a binary blend of the F2 2 2 , F224

and F226 fractions. The contribution of each fraction to the concentration of each

binary blend was equivalent to the screening concentration, 4 and 3 //g/cm , of the F22

fraction (for details see previous experiment). Only binary blends that were active at

the initial experiment were tested further.

147

Table 5.13 Percentage mortality of BO adults exposed to the F22 fraction and different combinations of the fractions obtained from the HPLC

fractionation using the Petri dish exposure bioassay.

Concentrationp g / c m 2

% MortalityTreatment Mean8 ± SEb, at hours after treatment

24 48 72

1st Experiment4.000 F22 46 ± 6.8a 68 ± 6.6a 68 ± 6.6a

0.364 + 0.364 + 0.892 F222 + F224 + F226 38 ± 6.6a 52 ± 8.0a 54 ± 6.8a0.756 + 0.720 + 0.376 + 0.892 + 0.520 F221 + F2 2 3 + F225 + F226 + F227 2 ± 2.0b 2 ± 2.0b 2 ± 2.0b

0.756 + 0.364 + 0.720 + 0.364 + 0.376 + 0.520 F22I + F222 + F223 + F224 + F225 + F227 0 ± 0.0b 0 ± 0.0b 0 ± 0.0b0.000 Control 0 ± 0.0b 2 ± 2.0b 2 ± 2.0b

2nd Experiment

3.000 F22 16 ± 4.0a 40 ± 6.3a 42 ± 5.8a0.273 + 0.273 + 0.669 F222 + F224 + F226 16 ± 5.1a 34 ± 6.8a 34 ± 6.8a

0.000 Control 0 ± 0.0b 0 ± 0.0b 0 ± 0.0b

"Mean o f 5 replicates o f 10 insects each (5 males and 5 females). bSE: Standard error o f mean. cTreated with 1 ml o f DCM plus olive oil (0.65 % v/v). The two experiments were not run simultaneously. Means within a column o f a single experiment followed by the same letter are not significantly different {P = 0.05, Tukey’s HSD test). Mortalities were transformed before ANOVA to arcsine V/>, where p is the proportion o f dead insects. Means ± SE o f untransformed data are reported.

The results of the two experiments are displayed in Table 5.14. The F222 + F224 and

F222 + F226 binary blends were inactive at the initial experiment, whereas the F22

fraction and the F224 + F226 binary blend were active and did not differ significantly at

both experiments at all mortality recording times.


binary blends of the three major peaks of the F22 fraction using the Petri dish exposure

bioassay.

Concentration//g/cm2 Treatment

% Mortality

Mean8 ± SE1b, at hours after treatment

24 48 72

1stExperiment4.000 F22 40 ± 5.5a 64 ± 5.1a 66 ± 5.1a

0.364 + 0.364 F222 + F224 34 ± 5.1a 56 ± 5.1a 56 ± 5.1a0.364 + 0.892 F222 + F226 0 ± 0.0b 2 ± 2.0b 2 ± 2.0b0.364 + 0.892 F224 + F226 0 ± 0.0b 0 ± 0.0b 2 ± 2.0b

0.000 Control0 2 ± 2.0b 4 ± 2.4b 4 ± 2.4b2ndExperiment

3.000 F22 20 ± 4.5a 34 ± 7.5a 36 ± 6.8a0.273 + 0.669 F 224 + F226 18 ± 3.7a 26 ± 6.0a 26 ± 6.0a

0.000 Control 2 ± 2.0b 2 ± 2.0b 2 ± 2.0b

“Mean o f 5 replicates o f 10 insects each (5 males and 5 females). bSE: Standard error o f mean. cTreated with 1 ml o f DCM plus olive oil (0.65 % v/v). The two experiments were not run simultaneously. Means within a column o f a single experiment followed by the same letter are not significantly different (P - 0.05, Tukey’s HSD test). Mortalities were transformed before ANOVA to arcsine V/>, where p is the proportion o f dead insects. Means ± SE o f untransformed data are reported.


that were eluted during HPLC fractionation to the F224 and F226 fractions. The activity

of the F22 fraction can fully be ascribed to the activity of the F224 + F226 binary blend.

Moreover, the chemical(s) present in the F222 fraction did not made any contribution

to the insecticidal activity of the F22 fraction.

149

In a final set of experiments the activity of the petroleum ether peel extract was

compared to the activity of a blend consisted of the Fi, F224 and F226 fractions, in order

to confirm that the activity of the petroleum ether peel extract was due to the presence

of these fractions in the extract. The contribution of each fraction to the concentration

of the blend was equivalent to the screening concentration of the petroleum ether peel

extract. In detail, at the screening concentration of 25 and 15 //g/cm2 the concentration

of the Fi fraction was 6.78 and 4.07 //g/cm2, the concentration of the F224 fraction was

0.35 and 0.21 //g/cm2, while the concentration of the F226 fraction was 0.85 and 0.51

//g/cm2, respectively. At this experiment olive oil was not added to treatment

solutions.

The results of these experiments are presented in Figure 5.15. There were no

significant differences in activity between the petroleum ether peel extract and the

blend of the Fi, F224 and F226 fractions at both concentrations at all mortality recording

times.

The results clearly show that the activity of the petroleum ether peel extract was the

result of the synergistic action among the Fi, F224 and F226 fractions. Since Fi fraction

consisted of a mixture of compounds it can be concluded that at least three

compounds contribute for the overall activity of the petroleum ether peel extract.

150


extract of CA and a blend consisted of the Fi, F224 and F226 fractions using the Petri

dish exposure bioassay.

Concentrationpg/cm2

% MortalityTreatment Mean® ± SEb, at hours after treatment

24 48 72 96

1st Experiment25 PEC 8 + 3.7a 60 ± 8.4a 78 ± 8.6a 80 + 7.1a

6.78 + 0.35 +0.85 Fl + F224 + F226 4 ± 2.4a 54 ± 8.7a 66+ 10.3a 66+ 10.3a0 Control*1 0 + 0.0a 2 ± 2.0b 2 ± 2.0b 2 ± 2.0b

2nd Experiment15 PE 2 ± 2.0a 22 ± 7.3a 38 ± 5.8a 40 + 5.5a

4.07 + 0.21+0.51 Fl + F224 + F226 2 ± 2.0a 24 + 5.1a 34 ± 4.0a 36 + 5.1a0 Control 0 ± 0.0a 0 ± 0.0b 0 ± 0.0b 0 ± 0.0b

“Mean of 5 replicates of 10 insects each (5 males and 5 females). bSE: Standard error of mean. Ê: Petroleum ether peel extract of CA. dTreated with DCM. The two experiments were not run simultaneously. Means within a column of a single experiment followed by the same letter are not significantly different (P = 0.05, Tukey’s HSD test). Mortalities were transformed before ANOVA to arcsine Vp, where p is the proportion of dead insects. Means ± SE of untransformed data are reported.

5.4 DISCUSSION

Bioassay-guided fractionation of the petroleum ether peel extract resulted in the

isolation of three pure compounds (fractions F222, F224 and F2 2 6). Only the F226 fraction

was toxic when tested individually. The F224 fraction was found to act synergistically

to the F226 fraction, whereas the F222 fraction did not show any significant activity.

The fractionation procedure of two sequential gravity silica columns followed by

HPLC for the final step of purification was proven sufficient to isolate the

insecticidally active chemicals present in the peel extract.

The amount of chemicals recovered in each fractionation step was considerably high

(lowest recovery 96.7 % for the first gravity column whereas for the second gravity

column and HPLC recovery was ca. 100 %), indicating that only a very small portion

of compounds were not eluted from the columns. TLC analysis on the fractions

obtained from the two gravity columns and HPLC analysis on the HPLC fractions

indicate that decomposition did not occur during the whole fractionation process and

that adequate separation was achieved in each fractionation step.

Bioassays conducted with the fractions obtained from the fractionation process

revealed that the activity of the petroleum ether peel extract prior to fractionation can

be fully explained by the synergistic activity among Fi, F224 and F2 26 fractions That

finding indicates that the peel extract did not loss its activity when passed through the

columns.

Synergism among several compounds, usually chemically related, is in most cases

accountable for the activity of plant extracts. Plant extracts have different biological

152

characteristics from pure compounds. Mixtures may be more effective due to

synergism or joint action of active compounds with different mechanisms (Assabgui

et al., 1997). Berenbaum et al. (1991) investigated whether a mixture of 6

furanocoumarins isolated from fruits of Pastinaca sativa (Apiaceae) is toxicologically

more effective against Heliothis zea larva than is an equivalent amount of

xanthotoxin, which is the major furanocoumarin present in the mixture. They found

that the mixture of six furanocoumarins was more effective than equivalent amount of

pure xanthotoxin in reducing growth of Heliothis zea larva.

Moreover, the evolution of resistance to plant extracts is extremely slow in contrast to

resistance to pure compounds which develops rapidly. Feng and Isman (1995) found

that peach aphid lines selected with pure azadirachtin, developed a 9-fold resistance to

this substance over 40 generations, but a parallel line selected with neem seed extracts

containing azadirachtin and other materials did not develop significant resistance. It

was suggested that the reason underlying the extremely slow evolution of resistance to

plant extracts is that it is more difficult for an insect to develop several adaptations

simultaneously (Brattsten et al., 1986).

McLaughlin et al. (1997) pointed out that by the use of crude, chemically unrefined,

plant extracts, containing mixtures of the bioactive plant components, rather than the

use of pure individual components, insect resistance is much less likely to develop,

the environmental load of individual components is lessened and the products of

crude extracts are cheaper to prepare. In addition, Weaver et al. (1997) stated that

plant extracts subjected to simple chemical fractionation procedures may have a great

potential for insect control.

153

Considering the advantages of using plant extracts and taking into account the

synergistic activity of the petroleum ether peel extract it can be concluded that crude

or partially purified petroleum ether peel extracts of CA could have great potential,

particularly in the developing countries, for the control of insect pests.

Having obtained reasonable quantities of fractions F2 2 2 , F224 and F226 (pure

compounds), the next step, which is presented in Chapter 6, is to fully characterise

these components using spectroscopic methods. Moreover, the activity of the

characterised compounds was compared with standards or synthesised compounds.

154

CHAPTER 6

IDENTIFICATION OF ISOLATED INSECTICIDALLY ACTIVE

COMPONENTS

6.1 INTRODUCTION

Generally, the methods employed for the determination of the structure of unknown

compounds can be divided into two categories: chemical and physicochemical

(spectroscopic). In essence, chemical methods involve the use of degradative

techniques to produce simpler and more readily analysed compounds from an

unknown. Success of these methods hinges on the ability to work with small

quantities of material, and the accurate observation and interpretation o f the results

obtained (Harwood et al., 1999). Addition reactions, reductive and oxidative

processes are the main chemical methods used (Stevens, 1998). However, nowadays

structure determination relies heavily on spectroscopic methods. The use of these

techniques has revolutionised structure organic chemistry, and their impact cannot be

overstated (Harwood et al., 1999). With the advent of modem chemistry, the

structures of many biologically active compounds became known, and the systematic

studies of natural products that protected plants from pests became a recognised

activity within the field of chemistry (Hedin and Hollingworth, 1997).

The most widely spectroscopic techniques used are: ultraviolet (UV) spectroscopy,

infrared (IR) spectroscopy, nuclear magnetic resonance (NMR) spectroscopy and

mass (MS) spectrometry (Williams and Fleming, 1995). With the exception of MS, all

155

other spectroscopic methods are non-destructive, thus permitting further examination

of the sample; fortunately MS requires the least quantity of sample (Houghton and

Raman, 1998). UV, IR and NMR come under the general heading of absorption

spectrometry while MS is fundamentally different to these in that it does not involve

absorption of radiation (Harwood et al., 1999). The uses of these methods are stated

by Williams and Fleming (1995). UV spectroscopy is used to detect conjugated

systems, because the promotion of electrons from the ground state to the excited state

of such systems gives rise to absorption in this region. ER is used to detect and

identify the vibrations of molecules, and specifically the characteristic vibrations of

the double and triple bonds present in many functional groups. NMR uses a longer

wavelength of the electromagnetic spectrum to detect changes in the alignment of

nuclear magnets in strong magnetic fields. MS measures the mass-to-charge ratio of

organic ions. Structural information comes from the moderately predictable

fragmentation organic molecules undergo; the masses of the fragment ions can often

be related to likely structures. However, it cannot be overemphasized that the method

of greatest importance in structure determination is NMR. As NMR increased in

power over the last three decades, UV and IR have decreased in importance.

In this chapter the structure of the isolated chemicals present in the petroleum ether

peel extract of CA was determined using the following spectroscopic methods: UV,

IR, MS and NMR. Moreover, biological and spectroscopic studies with isolated and

synthetic compounds were performed in order to confirm our results.

156

6.2 METHODS

6.2.1 Spectroscopic methods

The following spectroscopic methods were employed in order to elucidate the

chemical structure of the three isolated compounds from the petroleum ether peel

extract.

UV Spectroscopy

UV spectra of F2 22, F224, F226 fractions were obtained during the HPLC fractionation

of the F22 fraction of the petroleum ether peel extract (for details see section 5.2.1).

UV spectra were recorded in the region of 245-365 nm and printed (GP-100 Graphics

Printer, Perkin Elmer).

Gas chromatography - Fourier Transform Infrared Spectroscopy.

GC-FTIR analysis of F22 2 , F2 2 4 , F226 fractions was carried out on a Perkin Elmer

Autosystem gas chromatograph interfaced to a Perkin Elmer System 2000 (Perkin

Elmer Ltd., Buckinghamshire, England) Fourier transform infrared spectrometer

equipped with a liquid-nitrogen cooled narrow-band (4000 - 750 cm"1) infrared

detector (mercury cadmium telluride). Helium was used as the carrier gas (2 ml/min)

and the effluent from the GC was mixed with the sweep gas helium (0.5 ml/min;

transfer line temperature 280 °C) and passed through the IR light pipe with KBr

windows. All spectra were obtained at 8 cm ' 1 resolution. All samples were

chromatographed on a 30 m x 0.32 mm x 1 pm film thickness DB-5 column (J&W

Scientific). The oven temperature program was 50 °C for 2 min, then 5 °C/min to 280

°C and held for 60 min. Splitless injections were made (1 /A) at an injector

temperature of 250 °C and a splitless period of 90 sec.

157

Gas Chromatography - Mass Spectrometry.

Gas chromatography - mass spectrometry analysis of F2 2 2 , F22 4 , F2 26 fractions was

carried out on a Hewlett Packard 5890 Series II gas chromatograph interfaced to a

Fisons VG Trio 1000 (Manchester M23 9BE, UK) quadrupole mass spectrometer.

Electron impact ionization was used, with an electron energy of 70 eV and a trap

current of 200 pA. The interface was kept at 290 °C. All samples were

chromatographed on a 60 m x 0.25 mm x 0.1 pm film thickness DB-5 column (J&W

Scientific). The oven temperature was set at 50 °C for 2 min, then raised to 250 °C at

5 °C/min, held for 1 min, then 2 °C/min to 280 °C and held for 50 min. Helium was

used as the carrier gas at a flow rate of 1 ml/min. Splitless injections were made (1 pX)

at an injector temperature of 250 °C and a splitless period of 90 sec. 6',7'-

epoxybergamottin could not be chromatographed on the GC as it probably

decomposes on the hot injector temperature. For this compound, mass spectrometric

data were obtained by the use of the Direct Contact Inlet (DCI) probe of the VG Trio

1000 mass spectrometer. The DCI was set at 0 mA for 2 min, then raised to 200 mA

at 100 mA/min, held for 2 min, then 300 mA/min to 1000 mA held for 4 min and then

ramped down to 0 mA at 200 mA/min where it stayed for 5 min.

NMR Spectroscopy

NMR spectra of F22 2, F224, F226 fractions were recorded on Bruker Advance DPX-400

spectrometer. CDCI3 was used as solvent. Tetramethylsilane or residual solvent peaks

(e.g. CHCI3) were used as frequency standards. Coupling constants were determined

using the computer program Multiplet and are quoted in hertz (Hz). Values are

reported to 0.1 Hz, but have an uncertainty of ± ca. 0.3 Hz (at 400 MHz), due to the

158

digital resolution of the FID accumulation and Fourier transformation. lH NMR

spectra were simulated using RACCOON.

6.2.2 Synthesis of 6r,7-epoxybergamottin

20 mg bergamottin and 20 mg MCPBA, in 5 ml DCM, were left to stand for 3 h at

room temperature (20 - 22 °C) (Dreyer and Huey, 1973; Bellevue et al., 1997). The

DCM solution was extracted three times with H2O (5 ml each time) and the aqueous

layer was discarded. Na2S0 4 was added then to the DCM solution in order to absorb

the water remained. The DCM solution was filtered and concentrated. Purification of

the 6 7 '-epoxybergamottin was done by HPLC using the same chromatographic

conditions as in section 5.2.1.

6.2.3 Bioassay of isolated and synthetic compounds

The Petri dish exposure bioassay was used in order to compare the insecticidal

activity of the isolated and synthetic compounds against 2-3 d-old adults of BO. Stock

solutions were prepared in DCM, as were subsequent dilutions. Olive oil was added to

each treatment solution and control (DCM) so that each cm2 of the total inner surface

of the Petri dishes was layered with 0.1 ju\ of olive oil. Fresh test solutions were

prepared for each replicate. The syringe used for the application of samples was

thoroughly cleaned after the application of each sample. Mortalities were recorded

every 24 h for 3 d after which no further increase in mortality was recorded.

Estimation ofLCsos

Five replicates of 10 BO flies each (5 males and 5 females) which were conducted

using 5 different generations of flies, were employed for the 5 concentrations and the

159

control. The concentrations tested for each chemical were chosen on the basis of

preliminary range-finding studies. The concentrations of the isolated and synthesised

6 ',7'-epoxybergamottin tested were 1.7, 2.0, 2.3, 2.6 and 2.9 /*g/cm2. Rotenone, a

commercial plant-derived insecticide was used for comparison. The concentrations of

rotenone tested were 0.015, 0.020, 0.025, 0.030 and 0.035 /^g/cm2. Concentrations for

each replicate of each chemical were applied from the lowest to the highest.

Screening experiments

Five replicates of 10 BO flies each (5 males and 5 females) were employed for each

treatment and the control. The 5 replicates were conducted on 5 successive days,

utilising the same generation of flies. Different concentrations of screened treatments

were not run simultaneously as screening experiments at lower concentrations

included only the more active treatments.

160

6.3 RESULTS

6.3.1 Spectroscopic data of the isolated compounds

In the present work the isolated compounds were subjected to UV, IR, MS and ]H

NMR analysis. Considering the spectral data obtained and comparing them with that

present in the literature we assigned fraction F222 as osthol, fraction F224 as bergapten

and fraction F226 as 6',7'-epoxybergamottin. Their chemical structures are presented in

Figure 6.1.

10

o.

(1) (2)

10

2’O. 4’

10' 9’

(3)

Figure 6 . 1 Chemical structures of osthol (1), bergapten (2) and 6 ',7'-

epoxybergamottin (3).

161

Spectroscopic data of osthol (fraction F 2 2 2 )

UV Spectroscopy

The UV spectrum of osthol showed two absorption maxima (Figure 6.2). The

absorbance at 310 nm was considerably higher than that at 250 nm. This spectrum is

consistent to that reported by Hadadek et al. (1994) for osthol isolated from

Peucedanum (Apiaceae: Apioideae) species.

rtU

-O.0O1724 5 320 345

Figure 6.2 UV spectrum of osthol.

IR Spectroscopy

The most intense peak in the IR spectra of osthol (Figure 6.3) is the peak at 1770 cm' 1

that originates from the carbonyl absorption of the lactone moiety. Peaks at 1600 cm' 1

to 1450 cm' 1 are characteristic of the conjugated aromatic system. The C-0 stretching

of the lactone moiety and the absorptions of the =C-0-C group occur in the region of

1300 cm ' 1 to 1050 cm'1. The absorption at ca. 830 cm ' 1 is characteristic of the C-H out

of plane vibration of the aromatic ring the frequency of which is determined by the

number of adjacent hydrogen atoms of the ring. Finally the peaks in the region of

3000 cm*1 to 2800 cm' 1 are due to the absorption frequencies of the CH, CH2 and CH3

groups of the compounds.

162

100,0

3047,1498 2848,63

1399,42 829,082923,071490,87

96 1247,54

941605,44

1110,76

%T92

90

88

1769,92

85,632004000,0 3600 2800 2400 2000 1800 1600 1400 1200 1000 800 700,0

cm-1

Figure 6.3 IR spectrum of osthol.

Mass Spectroscopy

For osthol, the parent ion m/z 244 (100 %) is the base peak of the spectrum (Figure

6.3). Loss of the CH3 group probably occurs from the side chain and gives the peak

m/z 229 (77 %), and then the loss of CO gives the peak m/z 201 (72 %). The peak m/z

213 (48 %) is due to loss of the CH3O group from the molecular peak. Fission of the

/?-chain to the ring to remove a C4H7 radical gives the ion m/z 189 (81 %), which, after

hydrogen rearrangement, subsequently loses a CH20 group to give m/z 159 (40 %).

Finally, the expulsion of CO from the pyran ring gives the ion m/z 131 (62 %). This

spectrum is very similar to that reported by Zobel et al. (1991) for osthol isolated

from Apium graveolens and by Buiarelli et al. (1996) for osthol isolated from the non

volatile fraction of grapefruit essential oil.

163

244100 n

189229

201

131

213115

159

17512 £

103176

141160

148151132116107119 177

Figure 6.4 MS spectrum of osthol.

NMR Spectroscopy

5h (CDC13, 400 MHz ) 7.55 (d, 1H, J 9.4 Hz, 4-H), 7.20 (d, 1H, J 8 . 6 Hz, 5-H), 6.77

(d, 1H, J 8 . 6 Hz, 6 -H), 6.17 (d, lH ,./9 .4 Hz, 3-H), 5.16 (t, 1H, J7 .2 Hz, 2'-H), 3.85

(s, 3H, 6 '-H3), 3.45 (d, 2H, J 12 Hz, l'-H2), 1.78 (s, 3H, 4'-H3), 1.59 (s, 3H, 5'-H3).

The JH NMR spectrum of osthol is presented in Figure 6.5.

164

C S 8 « B K 8 2 ! S e ! B 8 = a i e S 8 » a 8 * l 5 E i * 8 5 8 8 5 S 8 R 5 S B « 8 * S S 8 » : s i ( * 8 5 l g s a 8 3 i 8 3 8 8 S * 8 - 8 E « a S i 8 S S S = l B S 8 S 8 8 B S 8

i i m m i ............................. '............ .........................................................................

J l

Figure 6.5 ]H NMR spectrum of osthol.

Spectroscopic data o f bergapten (fraction F224)

UV Spectroscopy

The UV spectrum of bergapten showed two absorption maxima (Figure 6.6). The

absorbance at 250 nm was considerably higher than that at 305 nm. This spectrum is

consistent to that reported by Hadacek et a l (1994) for bergapten isolated from

Peucedanum (Apiaceae: Apioideae) species.

hU

0 - . .

0 .0046220 245 270 95 320 345

Figure 6.6 UV spectrum of bergapten.

IR Spectroscopy

The gas phase IR spectra of bergapten (Figure 6.7) is very similar to that of osthol

since these two compounds possess common functional groups, such as the

conjugated aromatic system, the lactone moiety and the ether group, and consequently

they both exhibit the characteristic absorptions of these groups.

166

100,0

98

827,781349,76

96 1469,321582,31 1085,55

1630,75

%T

83,028004000,0 3800 3200 2400 2000 1800 1600 1400 1200 1000 800 700,0

cm-1

Figure 6.7 IR spectrum of bergapten.

Mass Spectroscopy

The parent ion m/z 216 (100 %) is the base peak of the mass spectrum of bergapten

(Figure 6 8 ). Loss of the CH3 radical from the parent ion gives the peak m/z 201

(35 %) whilst loss of the CO group leads to the formation of the peak m/z 188 (15 %).

The formed ion m/z 201 undergoes sequential losses of a CO group to give the peaks

at m/z 173 (64 %) and m/z 145 (33 %) respectively. This spectrum is consistent to that

reported by Valenciennes et al. (1999) for bergapten isolated from Euodia borbonica

(Rutaceae).

167

216100-1

173

145%

201

188217174

146

58m/z

Figure 6.8 MS spectrum of bergapten.

NMR Spectroscopy

5h (CDC13, 400 MHz) 8.14 (d, 1H, J 9.9 Hz, 4-H), 7.57 (d, 1H, J 2.4 Hz, 10-H), 7.11

(s, br, 1H, 8-H), 7.01 (dd, 1H, J 2.4 and 0.8 Hz, 9-H), 6.27 (d, 1H, J 9.9 Hz, 3-H),

4.27 (s, 3H, r-H 3). The JH NMR spectrum of bergapten is presented in Figure 6.9.

This spectrum is consistent to that reported by Sasaki et al. (1980) for bergapten

isolated from Glehnia littoralis (Umbelliferae).

168

rrrssrs aasjiSBBRSimSB

*28s ;«»esssa3sas58* = 8;a)E!88 = SiBesx!ss8i*gi*SSSi*«®**SS&sss88ssss*sas55*sstS 8 S 8 * 8 » S 8 2 S 8 8 s 9 ft S

W I W1 w

Figure 6.9 ]H NMR spectrum of bergapten.

Spectroscopic data o f 6\ 7'-epoxybergamottin (fraction F226)

UV Spectroscopy

The UV spectrum of 6',7'-epoxybergamottin was very similar to bergapten’s spectrum

(Figure 6 .10). The absorbance at 250 nm was considerably higher than that at 305 nm.

This spectrum is consistent to that reported by Dreyer and Huey (1973) for 6 ',7'-

epoxybergamottin isolated from Citrus macroptera.

rtU

0 . . .

195 220 24 270 320 345 37©

Figure 6.10 UV spectrum of 6',7'-epoxybergamottin.

Mass Spectroscopy

The El mass spectrum of 6',7'-epoxybergamottin does not exhibit the molecular ion at

m/z 354 as it rapidly fragments following the typical fragmentation pattern of the

epoxides with preferential cleavage alpha to the epoxy group. This cleavage gives rise

to the base peak of the spectrum at m/z 71 (100 %) corresponding to the C 4 H 7 O

fragment (Figure 6.11). The peak m/z 202 (69 %) originates probably from alpha

cleavage to the ether group after hydrogen rearrangement and charged retained at the

furanocoumarin moiety (loss of the CioHÔ group from the molecule of the 6 ',7'-

epoxybergamottin). The same cleavage, but this time with the charged retained at the

aliphatic moiety gives the peak at m/z 153 (30 %) corresponding to the C 10H 17O

170

fragment. Loss of the CO group from the m/z 202 ion gives the peak at m/z 174

(42 %). The peak m/z 145 (27 %) probably originates from the m/z 174 fragment with

loss of the CHO group.

100

202

174

118

^107 10J

1̂ ( ,;19 j 154 j

100 120 140 180 180 200 220 240 280 280 300 320 340 360 380 400

203135

Figure 6.11 MS spectrum of 6 7 -epoxybergamottin.

NMR Spectroscopy

5h (CDC13, 400 MHz) 8.09 (d, 1H, 79.9 Hz, 4-H), 7.53 (d, 1H, 72.5 Hz, 10-H), 7.09

(s, 1H, 8-H), 6.88 (m, 1H, 7 2.5 and 0.9 Hz, 9-H), 6.21 (d, 1H, J 9.9 Hz, 3-H), 5.52

(m, 1H, 2'-H), 4.89 (d, 2H, J 6.9 Hz, l'-H2), 2.64 (dd, 1H, J 7.0 and 5.4 Hz, 6'-H),

2.15 (m, 2H, 4'-H2), 1.67 (s, 3H, 10'-H3), 1.55 (m, 2H, 5'-H2), 1.25 (s, 3H, 9'-H3), 1.23

(s, 3H, 8'-H3). The ]H NMR spectrum of 6',7'-epoxybergamottin is presented in

Figure 6.12. This spectrum is consistent to that reported by Dreyer and Huey (1973)

for 6',7'-epoxybergamottin isolated from Citrus macroptera and by Ohta et al. (2002)

for 6',7'-epoxybergamottin isolated from grapefruit juice.

171

g 5 S S 8 a : ; S f 3 S 2 S i e S * R i S 1 l ! » R 8 S 5 ) B a 8 S 8 S 8 l l i 8 B 9 8 * * * a = 8 R 2 5 « B R i 8 8 s s S a S ! j £ | ? 8 f f ? S l B S 8 » s a S 3 £ S 2 5 * s 5 l ! S S S 3 ^ g

H n H M m H j m w r n m M

lL

f

w Jr zdLiL<1NJ

Figure 6.12 *H NMR spectrum of 6',7'-epoxybergamottin.

6.3.2 Synthesised 6',7-epoxy bergamottin

Selective epoxidation of the terminal bond of bergamottin gave the 6 ',7'-

epoxybergamottin in 69 % yield. Bellevue et al. (1997) obtained 6 ',7'-

epoxybergamottin in similar yield (65 %).

6.3.3 Spectroscopic data of synthetic osthol, bergapten and 6',7f-

epoxy bergamottin

All the spectroscopic data of the synthesised 6',7-epoxybergamottin, synthetic

bergapten purchased from Sigma and synthetic osthol kindly provided by May

Berenbaum, were identical to the isolated 6',7'-epoxybergamottin, bergapten and

osthol, respectively.\

6.3.4 Bioassays

Petri dish exposure bioassays were conducted in order to investigate whether there

were any difference in activity between isolated and synthetic 6',7'-epoxybergamottin,

bergapten and osthol.

Estimation ofLCsos

LC50S were estimated for the isolated and the synthesised in our laboratory 6 ',7-

epoxybergamottin in order to confirm that the observed activity of the isolated 6 ',7'-

epoxybergamottin was not due to minor compounds possibly present in it. Moreover,

LC50S of the botanical insecticide rotenone were estimated in order to compare its

activity to that of the 6',7'-epoxybergamottin.

173

A summary of probit analyses of mortality data for BO adults exposed to different

concentrations of the isolated and the synthetic 6',7'-epoxybergamottin and the

botanical insecticide rotenone using the Petri dish exposure bioassay is presented in

Table 6.1. LC50S were not estimated at 24 h after treatment as mortality at this time

was not significant high to estimate precisely the lethal concentrations. All P-values

of Table 6.1 were much greater than 0.05 and so Pearson’s goodness-of-fit tests

were not significant, indicating that the data fit the assumptions of the probit model

adequately.

Table 6.1 Summary of probit analyses of mortality data of BO exposed to different

concentrations of the isolated and the synthetic 6',7'-epoxybergamottin and the\

botanical insecticide rotenone using the Petri dish exposure bioassay.

h n LC50

(M g/cm2) 95 % CL Slope ± SE Intercept ± SE 7? df P

Isolated 6',7'-epoxybergamottin48 250 2.23 2.12-2.35 7.75 ± 1.09 2.29 ± 0.40 20.4 23 0.61572 250 2.19 2.07-2.31 7.53 ± 1.09 2.43 ± 0.39 20.4 23 0.617

Synthetic 6 7 '-epoxybergamottin48 250 2.38 2.27 - 2.53 7.33 ± 1.10 2.23 ±0.41 18.5 23 0.72872 250 2.35 2.23 - 2.48 7.62 ± 1.10 2.17 ±0.41 22.3 23 0.503

Rotenone48 250 0.025 0.023 - 0.027 5.88 ±0.75 14.42 ± 1.21 29.0 23 0.180

72 250 0.024 0.023 - 0.026 5.71 ±0.74 14.21 ± 1.19 27.7 23 0.228

Column headings are abbreviated as follows: h is the hours after treatment; n is the numbers of insects tested (mixed sexes) excluding the controls; 95 % CL is the 95 % confidence limits; Slope ± SE is the slope ± standard error; Intercept ± SE is the intercept ± standard error; X2 is the Pearson’s X2 goodness- of-fit test; df is the degrees of freedom; P is the probability.

Isolated and synthetic 6',7'-epoxybergamottin did not differ significantly as the 95 %

confidence limits overlap at both mortality recording times. Moreover, the insecticidal

activity of rotenone was at least 95 times greater than the activity of either isolated or

174

synthetic 6',7'-epoxybergamottin. That results confirm the insecticidal activity of the

6',7'-epoxybergamottin. However, its toxicity was substantially lower when compared

to the toxicity of the botanical insecticide rotenone. Moreover, it must be noticed that

either the isolated or the synthetic bergapten and osthol did not show in sign of

toxicity against BO adults even at 10 pg/cm2.

A very interesting finding was revealed when bergamottin was tested against BO

adults. Even at 10 jug/cm2 bergamottin did not exhibit any sign of activity. Since the

only difference between bergamottin (Figure 6.13) and 6',7'-epoxybergamottin (see

Figure 6.1) is the epoxide in the 6',7' position it can be concluded that the epoxide has

a critical role for the activity produced by the 6',7'-epoxybergamottin.

o.

Figure 6.13 The chemical structure of bergamottin.

Screening experiments

In order to confirm the synergistic action between 6',7'-epoxybergamottin and

bergapten a set of experiments with the following treatments was conducted: isolated

6',7'-epoxybergamottin plus isolated bergapten, isolated 6',7'-epoxybergamottin plus

synthetic bergapten, synthetic 6',7'-epoxybergamottin plus isolated bergapten and

175

synthetic 6',7'-epoxybergamottin plus synthetic bergapten. The contribution of each

compound to each blend was equivalent to 4 and 3 //g/cm2 of the F22 fraction. In

detail, the concentration of either isolated or synthetic bergapten tested was 0.364 and

0.273 while the concentration of either isolated or synthetic 6',7'-epoxybergamottin

tested was 0.892 and 0.669 /^g/cm2, in the initial and the subsequent experiment

respectively.

The results of these experiments (Table 6.2) reveal that synergism between 6',7'-

epoxybergamottin and bergapten occur only when the isolated bergapten was tested.

When the isolated bergapten was substituted by the synthetic bergapten synergism

was not evident indicating that other minor compounds present in the isolated\

bergapten must be responsible for the synergistic effects. Moreover, there was no

difference in activity when the isolated 6',7'-epoxybergamottin was substituted by the

synthetic 6',7'-epoxybergamottin confirming that the 6',7'-epoxybergamottin was

responsible for the observed synergism.

176

Table 6.2 Percentage mortality of BO adults exposed to different combinations of

isolated and synthetic 6',7'-epoxybergamottin and bergapten using the Petri dish

exposure bioassay.

Concentration p g/cm2

% MortalityTreatment“ Meanb ± SE°, at hours after treatment

24 48 72

0.892 + 0.364 IE + IB

1st Experiment 40 ± 4.5a 64 ± 5.1a 66 ± 6.0a

0.892 + 0.364 SE + IB 44 ± 4.0a 62 ± 5.8a 62 ± 5.8a

0.892 + 0.364 IE + SB 0 ± 0.0b 2 ± 2.0b 2 ± 2.0b

0.892 + 0.364 SE + SB 0 ± 0.0b 0 ± 0.0b 0 ± 0.0b

0.000 Control 0 ± 0.0b 0 ± 0.0b 0 ± 0.0b

0.669 + 0.273 IE + IB2nd Experiment

22 ± 5.8a 38 ± 8.6a 40 ± 7.1a

0.669 + 0.373 SE + IB 18 ± 3.7a 28 ± 5.8a 30 ± 5.5a0.000 Control 2 ± 2.0b 2 ± 2.0b 2 ± 2.0b

"IE: Isolated 6',7'-epoxybergamottin; SE: Synthetic 6',7'-epoxybergamottin; IB: Isolated bergapten; SB: Synthetic bergapten; Control: Treated with 1 ml of DCM plus olive oil (0.65 % v/v). '‘Mean of 5 replicates of 10 insects each (5 males and 5 females). CSE: Standard error of mean. The two experiments were not run simultaneously. Means within a column of a single experiment followed by the same letter are not significantly different (P = 0.05, Tukey’s HSD test). Mortalities were transformed before ANOVA to arcsine >Ip, where p is the proportion of dead insects. Means ± SE of untransformed data are reported.

177

6.4 DISCUSSION

Considering the data obtained from the UV, FTIR, MS and ]H NMR studies the

isolated pure compounds, fractions F2 22, F224 and F226 were assigned as osthol,

bergapten and 6',7'-epoxybergamottin, respectively. Comparison of our data to that

reported in the literature and authentic components revealed that the spectra were

identical, confirming our results.

The coumarin, osthol and the linear furanocoumarins or psoralens, bergapten and

6',7'-epoxybergamottin among other coumarins and furanocoumarins have been

recently isolated from the oxygen heterocyclic fraction of bitter orange essential oil by

normal-phase HPLC (Dugo et al., 1996). Coumarin and psoralen compounds are\

common to plants of the Rutaceae family (Gray and Waterman, 1978), to which

Citrus species belong. The coumarin and psoralen compounds present in the Citrus

peel oils are essentially non-volatile and, consequently, are not found in the distilled

oils (Stanley and Jurd, 1971). They exhibit strong absorption in the ultraviolet region

(Stanley and Jurd, 1971). Berenbaum (1991) reported that high performance liquid

chromatography has provided a fast, efficient, and sensitive method for separating and

quantifying coumarins and psoralens at the nanogram level by both normal- and

reverse-phase. In this study, these advantages during HPLC purification of osthol,

bergapten and 6',7'-epoxybergamottin were proven.

Since it is always possible that minor components can give rise to all or part of the

biological activity (Cannell, 1998) the activity of the isolated 6',7'-epoxybergamottin

was compared to the activity of 6',7'-epoxybergamottin synthesised in our laboratory.

It was revealed that the activity of the isolated and synthetic 6',7'-epoxybergamottin

178

did not differ significantly indicating that the 6',7'-epoxybergamottin was responsible

for the mortality of BO adults. This is the first report of the insecticidal activity of

6',7'-epoxybergamottin. However, another activity of 6',7'-epoxybergamottin has been

reported recently. Wangensteen et al. (2003) were found that 6',7'-epoxybergamottin

can inhibit the activity of the enzyme cytochrome P450 3A4 (CYP3 A4).

Comparison of the insecticidal activity of 6 ',7 -epoxybergamottin and bergamottin

showed that the epoxide is essential for the observed activity of 6',7'-

epoxybergamottin, since bergamottin was inactive even at considerably higher

concentrations than that used in the concentration-response experiments with 6',7-

epoxybergamottin. Although bergamottin did not show any significant activity againstt

BO in the Petri dish exposure bioassay, it has been reported that species of

Lepidoptera were deterred from feeding when bergamottin was incorporated into their

diets (Stevenson et al., 2003).

Comparison of 6',7-epoxybergamottin with the commercially available botanical

insecticide rotenone revealed that 6',7'-epoxybergamottin was considerably less toxic

than rotenone. However, it should be noted that any work dealing with natural

compounds as sources of new insecticides, should consider positive results as an

indication of potential in a chemical series, rather than trying to discover compounds

usable per se. Few plant-derived chemicals can be as effective as the modem

synthetic insecticides. Bioactive natural compounds should therefore be considered

models which can serve for the development of series of more efficient,

biodegradable, safer insecticides (Escoubas et al., 1994). Moreover, since plant-

179

derived chemicals are known to exhibit increased specificity, it is possible that 6',7'-

epoxybergamottin can exhibit greater activity against other insect pests.

Bioassays conducted with isolated and synthetic bergapten and 6',7'-

epoxybergamottin revealed that minor compounds present in the isolated bergapten

should be accounted for the synergistic action between bergapten and 6',7'-

epoxybergamottin. Moreover, it should be noted that isolated and synthetic bergapten

did not produce any sign of toxicity against BO by the Petri dish exposure bioassay

even at considerably higher concentrations than that used in the concentration-

response experiments with 6',7'-epoxybergamottin. However, the antifeedant activity

of bergapten is well documented (Escoubas et al., 1992; Berdegue et al., 1997;i

Calcagno et al., 2002; Stevenson et al., 2003).

Petri dish exposure bioassays conducted with isolated and synthetic osthol did not

exhibit any activity against BO even at considerably higher concentrations than that

used in the concentration-response experiments with 6',7'-epoxybergamottin.

However, osthol is known that posses antibacterial (Stanley and Jurd, 1971),

antimalarial (Harbome and Baxter, 1993), antimutagenic (Harbome and Baxter,

1993), cytostatic (Gawron and Glowniak, 1987) and cytotoxic (Yang et al., 2003)

activity.

180

CHAPTER 7

GENERAL DISCUSSION

Plants in general have the ability to synthesise compounds that, while having no

apparent function in primary metabolism, are physiologically active in insects and

other organisms, so providing plants with one of their most important defense

mechanisms. During recent years, increasing political and consumer pressures to

reduce the use of synthetic insecticides, and the decreasing efficacy of them due to the

development of resistance, have fuelled the search among plants for potentially useful

insecticides. Plant-derived insecticides are likely to be more rapidly degradable and

consequently more environmentally and toxicologically safe and more selective,

which accords with the requirements of IPM.

Citrus species are known to contain chemicals that exhibit different properties

(toxicity, deterrence, feeding deterrence) against insects. Studies undertaken so far on

the insecticidal potency of Citrus species have mainly concentrated on Citrus limon,

while no attention was given to Citrus aurantium.

The results of my research reveal that peels of C. aurantium contain chemicals with

insecticidal properties. Petri dish and topical application bioassays shown that a

petroleum ether peel extract obtained using the ultrasound extraction method was

toxic against Bactrocera oleae and Ceratitis capitata adults. Bioassay-guided

fractionation of this extract resulted in the isolation of three compounds which were

181

elucidated by spectroscopic methods and were assigned as osthol (coumarin),

bergapten and 6',7'-epoxybergamottin (furanocoumarins). However, only the 6',7'-

epoxybergamottin was toxic against B. oleae and its toxicity was the same to the

synthesised 6',7'-epoxybergamottin. It should be noticed that the insecticidal activity

of the petroleum ether peel extract can only be fully explained by the synergistic

action between the 6 ',7-epoxybergamottin and two or more compounds, which were

not isolated and identified in this project.

Coumarins are secondary plant metabolites which can be divided into three major

classes: simple hydroxycoumarins, furanocoumarins and pyranocoumarins. Simple

hydroxycoumarins and their glucosides are found regularly in higher plants, havingi

been recorded in ca. 100 families (Murray, 1991). The furano- and pyranocoumarins,

which usually occur in the free state in fruits and roots, are mainly restricted to two

large families, the Umbelliferae (Apiaceae) and the Rutaceae, where they occur

widely. They have been recorded occasionally in six other families: the Asteraceae,

Fabaceae, Moraceae, Pittosporaceae, Solanaceae and Thymeleaceae (Harbome and

Baxter, 1993). The site of storage or accumulation of furanocoumarins in plants is

also the site of synthesis, and especially for the rutaceous genus Citrus,

furanocoumarins are localised in oil glands in the peel (Berenbaum, 1991), as

revealed in this study.

Furanocoumarins are more biologically active than simple coumarins (Harbome and

Baxter, 1993). Their biological roles in their host plants have not been well

understood, but a likely function is to protect, through toxicity, plants from a wide

array of natural enemies (Guo and Yamazoe, 2004). This is supported by the

182

correlation of bacterial toxicity and insect resistance with higher concentrations of

furanocoumarins in celery and other furanocoumarin-containing plants (Trumble et

al., 1990; McCloud et al., 1992; Cianfrogna et al., 2002). Moreover, UV light induces

the biosynthesis of furanocoumarins, which are often produced in tissue exposed to

the sun. Their accumulation in high light environments may reflect their ability to

function as UV-absorbent screens, protecting plant ptotosynthetic apparatus from

damaging radiation (Berenbaum, 1991).

Furanocoumarins are unusual among plant secondary compounds because their toxic

properties are greatly enhanced in the presence of UV light. Photoactivity of

furanocoumarins is due to the formation of an excited triplet state on absorption of a

proton. The excited triplet state furanocoumarin can react directly with molecules

such as pyrimidine bases. Alternatively, the excited triplet state furanocoumarin can

react with ground state oxygen. By energy transfer, singlet oxygen can be generated,

or, by either electron or charge transfer, toxic oxyradicals such as hydroxy radical or

superoxide anion radicals can be formed. These molecules can then proceed to react

with molecules, including DNA, to cause toxicity (Berenbaum, 1991, 1995; Sastry et

al., 1997).

Furanocoumarins elicit behavioural responses and physiological effects in

herbivorous insects. They are repellents and feeding deterrents (Klocke et al., 1989;

Berdegue et al., 1997) as well as toxins (Berenbaum, 1978). In addition to their

inherent toxicity, they may act synergistically (Brewer et al., 1995) or antagonistically

(Diawara et al., 1993) when mixed with other furanocoumarins. Moreover, synergism

has also been observed in mixtures of the furanocoumarin xanthotoxin with the

183

nonfuranocoumarins myristicin, saffole, and fagaramide (Berenbaum and Neal, 1985;

Neal, 1989). Toxicity in most instances is greatly enhanced in the presence of UV

light, suggesting that photoactivation, possibly involving DNA, is a major mechanism

of toxicity (Berenbaum, 1991). However, there are also toxic effects of

furanocoumarins that are independent of UV light (Berenbaum, 1978). Insects feeding

on plants with high contents of furanocoumarins possess adaptations for reducing the

toxicity of furanocoumarins. Many associates of furanocoumarin-containing plants

avoid light and thereby avoid the photoactivating effects of UV on furanocoumarins

(Berenbaum, 1990). Moreover, in some insects a biochemical resistance to

furanocoumarins has been evolved (Hung et al., 1995).

Furanocoumarins a,re toxic against a wide variety of organisms, including bacteria,

fungi, plants, nematodes, molluscs, fish, birds and mammals (Berenbaum, 1991).

Moreover, some linear furanocoumarins disrupt the detoxification capability of

organisms by acting as suicide substrates for cytochrome P-450 monooxygenases, an

enzyme system involved in detoxification processes in mammals (Koenigs and

Trager, 1998) and insects (Neal and Wu, 1994).

Phytophotodermatitis, an epidermal reaction to furanocoumarins in the presence of

UV light, is the most widely recognised manifestation of furanocoumarin activity. In

all organisms afflicted, phytophotodermatitis manifests itself in the form of erythema,

bullous eruptions, and pigmentation at the point of contact with furanocoumarin-rich

plant tissue (Fleming, 1990). These symptoms are well known from celery workers

and the causal agents are the linear furanocoumarins psoralen, 5-methoxypsoralen

(bergapten), and 8-methoxypsoralen (xanthotoxin) (Austad and Kavli, 1983;

184

Ashwood-Smith et a l, 1985). Human food plants known to cause photodermatitis

include carrots, celery, parsnip, dill, fennel, caraway, parsley, lovage, anise, and

chervil in the Apiaceae; figs in the Moraceae; and lemon, orange, and grapefruit in the

Rutaceae (Murray et al., 1982).

Furanocoumarins have been used since ancient times to treat human skin disorders

such as skin depigmentation (vitiligo), psoriasis, mycosis fungoides, polymorphous

photodermatitis and eczema (Musajo and Rodighiero, 1962; Scott et al., 1976).

However, the use of these compounds in medicine has been associated with higher

incidence of skin cancer (Musajo and Rodighiero, 1962; Stem et al., 1979; Grekin and

Epstein, 1981). Moreover, furanocoumarins have been reported to be carcinogenic

and mutagenic (Roelandts, 1984). Therefore, benefit and risk should be carefully

balanced against each other when furanocoumarins are used in medicine (Daniel et

a l, 1999).

Citrus oils are used in the formulation of perfumes and cosmetics, and coumarins are

used for their flavour quality and are added to sunscreen formulations to enhance

tanning induced by ultraviolet radiation (Mabry and Ulubelen, 1980). Moreover,

several plants containing furanocoumarins, e.g. lemon, celery, parsnip, parsley and

carrots, are part of the human diet (Daniel et al., 1999). Based on the average

consumption and furanocoumarin contents of different food items, the uptake of

furanocoumarins in the United States was calculated to be 1.3 mg per person per day

(Wagstaff, 1991). It can be concluded that the carcinogenic potential of

furanocoumarins is low, although their combination with exposure to UV radiation

could pose a hither to unquantified risk.

185

Further studies will need to be done in order that the full insecticidal potential

usefulness of C. aurantium can be explored. Different stages of insects from a broad

range of families should be tested using different bioassay methods. Since botanical

insecticides are generally species-specific and also different bioassays produce

different results, it is highly likely that a more pronounced activity will be shown.

Even fractions, extracts and isolated compounds that were inactive by the bioassays

and insect species used in this study, can be revealed to be veiy promising as pest

control chemicals. Moreover, further work is needed in order to determine the

chemical structure of the synergistic chemicals present in the peel extract. Besides

toxicity studies, it is also important to determine other factors such as persistence and

effects on man and other non-target organisms.

Roark (1942) suggested that the reasons for seeking new natural insecticides included

the continuing need to combat insects which had not yielded to effective control, to

eliminate chemicals poisonous to man, to find substitutes for existing chemicals, and

to provide a supply to meet insecticidal shortages. These reasons appear as valid today

as they did then, despite the dominance of synthetic insecticides, many of which are

now environmental hazards.

186

CHAPTER 8

REFERENCES

Abbassy, M.A.A., Hosny, A.H., Lamaei, O. and Choukri, O. (1979) Insecticidal and

synergistic Citrus oils isolated from Citrus peels. Mededelingen van de

Landbouwwetenschappen Rijksuniversiteit Gent, 44, 21-9.

Abbott, W.S. (1925) A method of computing the effectiveness of an insecticide.

Journal o f Economic Entomology, 18, 265-7.

Adan, A., del Estal, P., Budia, F., Gonzalez, M. and Vinuela, E. (1996) Laboratory

evaluation of the novel naturally derived compound spinosad against Ceratitis

capitata. Pesticide Science, 48,261-8.

Ahn, Y.J., Know, M., Park, H.M. and Han, C.K. (1997) Potent insecticidal activity of

Ginkgo biloba derived trilactone terpenes against Nilaparvata lugeus, in

Phytochemicals for Pest Control (eds P.A. Hedin, R.M. Hollingworth, E.P.

Masler, J. Miyamoto and D.G. Thompson), ACS Symposium Series No. 658,

American Chemical Society, Washington, pp. 90-105.

Alberola, T.M., Aptosoglou, S., Arsenakis, M., Bey, Y., Delrio, G., Ellar, D.J., Ferre,

J., Granero, F., Guttmann, D.M., Koliais, S., Martinez-Sebastian, M.J., Prota,

R., Rubino, S., Satta, A., Scarpellini, G., Sivropoulou, A. and Vasara, E.

(1999) Insecticidal activity of strains of Bacillus thuringiensis on larvae and

adults of Bactrocera oleae Gmelin (Dipt.: Tephritidae). Journal o f

Invertebrate Pathology, 74, 127-36.

Albrecht, C.P. and Sherman, M. (1987) Lethal and sublethal effects of avermectin Bi

on three fruit fly species (Diptera: Tephritidae). Journal o f Economic

Entomology, 80, 344-7.

Alford, A.R., Cullen, J.A., Storch, R.H. and Bentley, M.D. (1987) Antifeedant

activity of limonin against the Colorado potato beetle (Coleoptera:

Chrysomelidae). Journal o f Economic Entomology, 80, 575-8.

187

Amason, J.T., Philogene, B.J.R., Donskov, N. and Kubo, I. (1987) Limonoids from

Meliaceae and Rutaceae reduce feeding, growth and development of Ostrinia

nubilalis. Entomologia Experimental is et Applicata, 43, 221-6.

Arnold, A.J. (1967) Hand-operated micro-applicator to deliver drops of five sizes.

Laboratory Practice, 16, 56-7.

Ashwood-Smith, M.J., Ceska, O. Chaudhary, S.K. (1985) Mechanism of

photosensitivity reactions to diseased celery. British Medical Journal, 290,

1249.

Assabgui, R., Lorenzetti, F., Terradot, L., Regnault-Roger, C., Malo, N., Wiriyachitra,

P., Sanchez-Vindas, P.E., San Roman, L., Isman, M.B., Durst, T. and

Amason, J.T. (1997) Efficacy of botanicals ffom the Meliaceae and

Piperaceae, in Phytochemicals for Pest Control (eds P.A. Hedin, R.M.

Hollingworth, E.P. Masler, J. Miyamoto and D.G. Thompson), ACS

Symposium Series No. 658, American Chemical Society, Washington, pp. 38-

48.

Austad, J. and Kavli, G. (1983) Phototoxic dermatitis caused by celery infected by

Sclerotinia sclerotiorum. Contact Dermatitis, 9, 448-51.

Back, E.A. and Pemberton, C.E. (1915) Susceptibility of Citrus fruits to the attack of

the Mediterranean fruit fly. Journal o f Agricultural Research, 3, 311-30.

Back, E.A. and Pemberton, C.E. (1918) The Mediterranean fruit fly in Hawaii. United

States Department o f Agriculture Technical Bulletin, 640, 1-43.

Baker, R., Herbert, R.H., Howse, P.E., Jones, O.T., Francke, W. and Reith, W. (1980)

Identification and synthesis of the major sex pheromone of the olive fly,

Dacus oleae. Journal o f the Chemical Society, Chemical Communications, 1,

52-3.

Balandrin, M.F., Klocke, J.A., Wurtele, E.S. and Bollinger, W.H. (1985) Natural plant

chemicals: Sources of industrial and medicinal materials. Science, 228, 1154-

60.

Bellevue, F.H., Woster, P.M., Edwards, D.J., He, K. and Hollenberg, P.F. (1997)

Synthesis and biological evaluation of 6,7-dihydroxybergamottin (6,7-DHB), a

188

naturally occurring inhibitor of cytochrome P450 3A4. Bioorganic &

Medicinal Chemistry Letters, 7, 2593-8.

Benner, J.P. (1993) Pesticidal compounds from higher plants. Pesticide Science, 39,

95-102.

Berdegue, M., White, K.K. and Trumble, J.T. (1997) Feeding deterrence of

Spodoptera exigua (Lepidoptera: Noctuidae) larvae by low concentrations of

linear furanocoumarins. Environmental Entomology, 26, 912-9.

Berenbaum, M. (1978) Toxicity of a furanocoumarin to armyworms: A case of

biosynthetic escape from insect herbivores. Science, 201, 532-4.

Berenbaum, M.R. (1990) Evolution of specialization in insect Umbellifer associations.

Annual Review o f Entomology, 35, 319-43.

Berenbaum, M.R. (1991) Coumarins, in Herbivores: Their Interactions with

Secondary Plant Metabolites, 2nd edition (eds G.A. Rosenthal and M.R.

Berenbaum), Vol. 2: The Chemical Participants, Academic Press, New York,

pp. 221-49.

Berenbaum, M. (1995) Phototoxicity of plant secondary metabolites: Insect and

mammalian perspectives. Archives o f Insect Biochemistry and Physiology, 29,

119-34.

Berenbaum, M. and Neal, J.J. (1985) Synergism between myristicin and xanthotoxin,

a naturally cooccuring plant toxicant. Journal o f Chemical Ecology, 11, 1349-

58.

Berenbaum, M.R., Nitao, J.K. and Zangerl, A.R. (1991) Adaptive significance of

furanocoumarin diversity in Pastinaca sativa (Apiaceae). Journal o f Chemical

Ecology, 17, 207-15.

Bocco, A. Cuvelier, M., Richard, H. and Berset, C. (1998) Antioxidant activity and

phenolic composition of Citrus peel and seed extracts. Journal o f Agricultural

and Food Chemistry, 46, 2123-9.

Boiler, E.F. (1987) Genetic control, in Integrated Pest Management (eds A.J. Bum,

T.H. Coaker and P.C. Jepson), Academic Press, London, pp. 161-87.

189

Brattsten, L.B., Holyoke, Jr.C.W., Leeper, J.R. and RafFa, K.F. (1986) Insecticide

resistance: Challenge to pest management and basic research. Science, 231,

1255-60.

Brewer, M.J., Meade, T. and Trumble, J.T. (1995) Development of insecticide-

resistant and -susceptible Spodoptera exigua (Lepidoptera: Noctuidae)

exposed to furanocoumarins found in celery. Environmental Entomology, 24,

392-401.

Broumas, T. and Haniotakis, G.E. (1994) Comparative field studies of various traps

and attractants of the olive fruit fly, Bactrocera oleae. Entomologia

Experimental is et Applicata, 73, 145-50.

Budia, F. and Vinuela, E. (1996) Effects of cyromazine on adult C. capitata (Diptera:

Tephritidae) on mortality and reproduction. Journal o f Economic Entomology,

89, 826-31.

Bueno, A.M. and Jones, O. (2002) Alternative methods for controlling the olive fly,

Bactrocera oleae, involving semiochemicals, in Use o f Pheromones and other

Semiochemicals in Integrated Control (eds P. Witzgall, B. Mazomenos and M.

Konstantopoulou), IOBC/WPRS Bulletin, 25, 147-56.

Buiarelli, F., Cartoni, G.P., Coccioli, F. and Leone, T. (1996) Analysis of bitter

essential oils from orange and grapefruit by high-performance liquid

chromatography with microbore columns. Journal o f Chromatography A , 730,

9-16.

Caccioni, D.R.L., Guizzardi, M., Biondi, D.M., Renda, A. and Ruberto, G. (1998)

Relationship between volatile components of Citrus fruit essential oils and

antimicrobial action on Penicillium digitatum and Penicillium italicum.

International Journal o f Food Microbiology, 43, 73-9.

Calapai, G., Firenzuoli, F., Saitta, A., Squadrito, F., Arlotta, M.R., Costantino, G. and

Inferrera, G. (1999) Antiobesity and and cardiovascular toxic effects of Citrus

aurantium extracts in the rat: A preliminary report. Fitoterapia, 70, 586-92.

Calcagno, M.P. Coll, J., Lloria, J., Faini, F. and Alonso-Amelot, M.E. (2002)

Evaluation of synergism in the feeding deterrence of some furanocoumarins

on Spodoptera littoralis. Journal o f Chemical Ecology, 28, 175-91.

190

Cannell, R.J.P. (1998) How to approach the isolation of a natural product, in Natural

Products Isolation (ed. R. J.P. Cannell), Methods in Biotechnology 4, Humana

Press, Totowa, pp. 1-51.

Champagne, D.E., Koul, O., Isman, M.B., Scudder, G.G.E. and Towers, G.H.T.

(1992) Biological activity of limonoids from the Rutales. Phytochemistry, 31,

377-94.

Choo, L.E.W., Tang, C.S., Pank, F.Y. and Ho, S.H. (2000) Comparison of two

bioassay methods for determining deltamethrin resistance in German

cockroaches (Blattodea: Blattellidae). Journal o f Economic Entomology, 93,

905-10.

Cianfrogna, J.A., Zangerl, A.R. and Berenbaum, M.R. (2002) Effects of

furanocoumarins on feeding behavior of parsnip webworms Depressaria

pastinacella. Journal o f Chemical Ecology, 28, 1365-75.

Coaker, T.H. (1987) Cultural methods: The crop, in Integrated Pest Management (eds

A.J. Bum, T.H. Coaker and P.C. Jepson), Academic Press, London, pp. 69-88.

Coats, J.R. (1994) Risks from natural versus synthetic insecticides. Annual Review o f

Entomology, 39,489-515.

Copping, L.G. and Hewitt, H.G. (1998) Chemistry and Mode o f Action o f Crop

Protection Agents, The Royal Society of Chemistry, Cambridge.

Dadang, Ohsawa, K., Kato, S. and Yamamoto, I. (1996) Insecticidal compound in

tuber of Cyperus rotundus L. against the diamondback moth larvae. Journal o f

Pesticide Science, 21, 444-6.

Dahm, P. A., Gurland, J., Lee, I. and Berlin, J. (1961) A comparison of some house fly

bioassay methods. Journal o f Economic Entomology, 54, 343-7.

Daniel, O., Meier, M.S., Schlatter, J. and Frischknecht, P. (1999) Selected phenolic

compounds in cultivated plants: Ecologic functions, health implications, and

modulation by pesticides. Environmental Health Perspectives, 107, 109-14,

Supplement 1.

da Silva Branco, E., Vendramin, J.D. and Denardi, F. (2000) Resistencia as Moscas-

das-frutas em Fruteiras, in Moscas-das-frutas de Importancia Econdmica no

191

Brasil (eds A. Malavasi and R.A. Zucchi), Holos Editora, Riberao Preto, pp.

161-7.

Dent, D. (1991) Insect Pest Management, CAB International, Wallingford.

Dent, D.R. (1995) Control measures, in Integrated Pest Management (ed. D. Dent),

Chapman & Hall, London, pp. 47-85.

Diawara, M.M., Trumble, J.T., White, K.K., Carson, W.G. and Martinez, L.A. (1993)

Toxicity of linear furanocoumarins to Spodoptera exigua: Evidence for

antagonistic interactions. Journal o f Chemical Ecology, 19, 2473-84.

Don-Pedro K.N. (1985) Toxicity of some Citrus peels to Dermestes maculates Deg.

and Callosobruchus maculatus (F). Journal o f Stored Product Research, 21,

31-4.

Don-Pedro, K.N. (1996) Fumigant toxicity is the major route of insecticidal activity of

Citrus peel essential oils. Pesticide Science, 46, 71-8.

Dreyer, D.L. and Hpey, P.F. (1973) Coumarins of Citrus macroptera. Phytochemistry,

12,3011-3.

van Driesche, R.G. and Bellows, T.S.Jr. (1996) Biological Control, Chapman & Hall,

New York.

Duffus, J.H. (1993) Glossary for chemists of terms used in toxicology. Pure and

Applied Chemistry, 65, 2003-122.

Dugo, P., Mondello, L., Cogliandro, E., Verzera, A. and Dugo, G. (1996) On the

genuineness of Citrus essential oils. 51. Oxygen heterocyclic compounds of

bitter orange oil (Citrus aurantium L.). Journal o f Agricultural and Food

Chemistry, 44, 544-9.

Escoubas, P., Lajide, L. and Mizutani, J. (1994) Insecticidal and antifeedant activities

of plant compounds: Potential leads for novel pesticides, in Natural and

Engineered Pest Management Agents (eds P. A. Hedin, J.J. Menn and R.M.

Hollingworth), ACS Symposium Series No. 551, American Chemical Society,

Washington, pp. 162-71.

192

Escoubas, P., Fukushi, Y., Lajide, L. and Mizutani, J. (1992) A new method for fast

isolation of insect antifeedant compounds from complex mixtures. Journal o f

Chemical Ecology, 18, 1819-32.

Ezeonu, F.C., Chidume, G.I. and Udedi, S.C. (2001) Insecticidal properties of volatile

extracts of orange peels. Bioresource Technology, 76, 273-4.

Feng, R. and Isman, M.B. (1995) Selection for resistance to azadirachtin in the green

peach aphid, Myzus persicae. Experientia, 51, 831-3.

Finney, D.J. (1971) Probit Analysis, 3rd edition, Cambridge University Press,

London.

Fleming, D. (1990) Dermatitis in grocery workers associated with high natural

concentrations of furanocoumarins in celery. Allergy Proceedings, 11, 125-7.

Forrester, N.W., Cahill, M., Bird, L.J. and Layland, J.K. (1993) Management of

pyrethroid and endosulfan resistance in Helicoverpa armigera (Lepidoptera:

Noctuidae) in Australia. Bulletin o f Entomological Research, 1-132,

Supplement 1.

Gallo, L.G., Allee, L.L. and Gibson, D.M. (1996) Insecticidal effectiveness of

Mammea americana (Guttiferae) extracts on larvae of Diabrotica virgifera

virgifera (Coleoptera: Chrysomelidae) and Trichoplusia ni (Lepidoptera:

Noctuidae). Economic Botany, 50, 236-42.

Gawron, A. and Glowniak, K. (1987) Cytostatic activity of coumarins in vitro. Planta

Medica, 6, 526-9.

Gbewonyo, W.S.K. (1991) Insecticidal actions of Piper guineense, University of

Birmingham, Ph.D. Thesis.

Gbewonyo, W.S.K. and Candy, D.J. (1992a) Separation of insecticidal components

from an extract of the roots of male Piper guineense (West African black

pepper) by gas chromatography. Toxicon, 30, 1037-42.

Gbewonyo, W.S.K. and Candy, D.J. (1992b) Chromatographic isolation of

insecticidal amides from Piper guineense roots. Journal o f Chromatography,

607, 105-11.

193

Gbewonyo, W.S.K., Candy, D.J. and Anderson, M. (1993) Structure-activity

relationships of insecticidal amides from Piper guineense root. Pesticide

Science, 37, 57-66.

Graves, J.B., Leonard, B.R. and Ottea, J.A. (1999) Chemical approaches to managing

arthropod pests, in Handbook o f Pest Management (ed. J.R. Ruberson), Marcel

Dekker, New York, pp. 449-86.

Gray, A.I. and Waterman, P.G. (1978) Coumarins in the Rutaceae. Phytochemistry,

17, 845-64.

Greany, P.D., Styer, S.C., Davis, P.L., Shaw, P.E. and Chambers, D.L. (1983)

Biochemical resistance of Citrus to fruit flies. Demonstration and elucidation

of resistance to the Caribbean fruit fly, Anastrepha suspensa. Entomologia

Experimentalis et Applicata, 34,40-50.

Greany, P.D., McDonald, R.E., Schroeder, W.J., Shaw, P.E., Aluja, M. and Malavasi,

A. (1994) Use of gibberellic acid to reduce Citrus fruit susceptibility to fruit

flies, in Bioregulators fo r Crop Protection and Pest Control (ed. P. A. Hedin),

ACS Symposium Series No. 557, American Chemical Society, Washington,

pp. 39-48.

Grekin, D.A. and Epstein, J.H. (1981) Psoralens, UVA (PUVA) and

photocarcinogenesis. Photochemistry and Photobiology, 33, 957-60.

Gullan, P.J. and Cranston, P. S. (1994) The Insects: An Outline o f Entomology,

Chapman & Hall, London.

Guo, L.Q. and Yamazoe, Y. (2004) Inhibition of cytochrome P450 by

furanocoumarins in grapefruit juice and herbal medicines. Acta

Pharmacologica Sinica, 25, 129-36.

Hadacek, F., Muller, C., Wemer, A., Greger, H. and Proksch, P. (1994) Analysis,

isolation and insecticidal activity of linear furanocoumarins and other

coumarin derivatives from Peucedanum (Apiaceae: Apioideae). Journal o f


Harbome, J.B. and Baxter, H. (1993) Phytochemical Dictionary. A Handbook o f

Bioactive Compounds from Plants, Taylor & Francis, London.

194

Harwood, L.M., Moody, C.J. and Percy, J.M. (1999) Experimental Organic

Chemistry: Standard and Microscale, 2nd edition, Blackwell Science, Oxford.

Hedin, P. A. and Hollingworth, R.M. (1997) New applications for phytochemical pest-

control agents, in Phytochemicals fo r Pest Control (eds P.A. Hedin, R.M.



12.

Hedin, P.A., Hollingworth, R.M., Masler, E.P., Miyamoto, J. and Thompson, D.G.

(1997) Preface, in Phytochemicals for Pest Control (eds P.A. Hedin, R.M.


Symposium Series No. 658, American Chemical Society, Washington.

Hendrichs, J. (1996) Action programs against fruit flies of economic importance:

Session overview, in Fruit Fly Pests: A World Assessment o f their Biology and

Management (eds B.A. McPheron and G.J. Steck), St. Lucie Press, Florida,

pp. 513-9. -

Hendrichs, J., Franz, G. and Rendon, P. (1995) Increased effectiveness and

applicability of the sterile insect technique through male only releases for

control of Mediterranean fruit flies during fruiting seasons. Journal o f Applied


Hendrichs, J., Robinson, A.S., Cayol, J.P. and Enkerlin, W. (2002) Medfly area wide

sterile insect technique programmes for prevention, suppression or eradication:

The importance of mating behavior studies. Florida Entomologist, 85, 1-13.

Hill, D.S. (1997) The Economic Importance o f Insects, Chapman & Hall, London.

Hinkle, N.C., Sheppard, D.C. and Nolan, Jr.M.P. (1985) Comparing residue exposure

and topical application techniques for assessing permethrin resistance in house

flies (Diptera: Muscidae). Journal o f Economic Entomology, 78, 722-4.

Hostettmann, K. and Wolfender, J. (1997) The search for biologically active

secondary metabolites. Pesticide Science, 51, 471-82.

Houghton, P.J. and Raman, A. (1998) Laboratory Handbook fo r the Fractionation o f

Natural Extracts, Chapman & Hall, London.

195

Hu, X.P. and Procopy, R.J. (1998) Lethal and sublethal effects of imidacloprid on

apple maggot fly, Rhagoletispomonella Walsh (Dipt., Tephritidae). Journal o f

Applied Entomology, 122, 37-42.

Hung, C., Prapaipong, H., Berenbaum, M.R. and Schulers, M.A. (1995) Differential

induction of cytochrome P450 transcrips in Papilio polyxenes by linear and

angular furanocoumarins. Insect Biochemistry and Molecular Biology, 25, 89-

99.

Isman, M.B., Gunning, P.J. and Spollen, K.M. (1997) Tropical timber species as

sources of botanical insecticides, in Phytochemicals fo r Pest Control (eds P.A.

Hedin, R.M. Hollingworth, E.P. Masler, J. Miyamoto and D.G. Thompson),

ACS Symposium Series No. 658, American Chemical Society, Washington,

pp. 27-37.

Iwuala, M.O.E., Osisiogu, I.U.W. and Agbakwuru, E.O.P. (1981) Dennetia oil, a

potential new insecticide: Tests with adults and nymphs of Periplaneta

americana and Zonocerus variegates. Journal o f Economic Entomology, 74,

249-52.

Jacobson, M. (1989) Botanical pesticides: Past, present, and future, in Insecticides o f

Plant Origin (eds J.T. Amason, B.J.R. Philogene and P. Morand), ACS


10.

Jayaprakasha, G.K., Singh, R.P., Pereira, J. and Sakariah, K.K. (1997) Limonoids

from Citrus reticulata and their moult inhibiting activity in mosquito Culex

quinquefasciatus larvae. Phytochemistry, 44, 843-6.

Jefferies, P.R., Toia, R.F., Brannigan, B., Pessah, I. and Casida, J.E. (1992) Ryania

insecticides: Analysis and biological activity of 10 natural ryanoids. Journal o f

Agricultural and Food Chemistry, 40, 142-6.

Jimenez-Peydro, R., Gimeno-Martos, C., Lopez-Ferrer, J., Serrano-Delgado, C. and

Moreno-Mari, J. (1995) Effects of the insect growth regulator cyromazine on

the fecundity, fertility and offspring development of Mediterranean fruit fly,

Ceratitis capitata Wied. (Dipt., Tephritidae). Journal o f Applied Entomology,

119,435-8.

196

Johansen, C. (1962) Principles of insect control, in Fundamentals o f Applied

Entomology (ed. R.E. Pfadt), The Macmillan Company, New York, pp. 138-

58.

Jones, O.T. (1998) Practical applications of pheromones and other semiochemicals, in

Insect Pheromones and their Use in Pest Management (eds P.E. Howse, I.D.R.

Stevens and O.T. Jones), Chapman & Hall, London, pp. 263-356.

Kadir, H.A., Zakaria, M.B., Kechil, A.A. and Azirum, M.S. (1989) Toxicity and

electrophysiological effects of Spilanthes acmella Murr. extracts on

Periplaneta americana L.. Pesticide Science, 25, 329-35.

Kapatos, E.T. (1989) Integrated pest management systems o f Dacus oleae, in Fruit

Flies: Their Biology, Natural Enemies and Control, Vol. 3B (eds A. S.

Robinson and G. Hooper), Elsevier, Amsterdam, pp. 391-8.

Kapatos, E.T., Fletcher, B.S., Pappas, S. and Laudeho, Y. (1977) The release of Opius

concolor and O. concolor var. siculus (Hym.: Braconidae) against the spring

generation t)f Dacus oleae (Dipt.: Trypetidae) on Corfu. Entomophaga, 22,

265-70.

Katsoyannos, B.I. and Papadopoulos, N.T. (2004) Evaluation of synthetic female

attractants against Ceratitis capitata (Diptera: Tephritidae) in sticky coated

spheres and McPhail type traps. Journal o f Economic Entomology, 97, 21-6.

Katsoyannos, P. (1992) Olive Pests and their Control in the Near East, FAO Plant

Production and Protection Paper No. 115, Food and Agriculture Organisation

of the United Nations, Rome.

Klocke, J.A. and Kubo, I. (1982) Citrus limonoid by-products as insect control agents.

Entomologia Experimental is et Applicata, 32, 299-301.

Klocke, J.A., Balandrin, M.F., Bamby, M.A. and Yamasaki, R.B. (1989) Limonoids,

phenolics, and furanocoumarins as insect antifeedants, repellents and growth

inhibitory compounds, in Insecticides o f Plant Origin (eds J.T. Amason,

B.J.R. Philogene and P. Morand), ACS Symposium Series No. 387, American

Chemical Society, Washington, pp. 136-49.

Koenigs, L.L. and Trager, W.F. (1998) Mechanism-based inactivation of P450 2A6

by furanocoumarins. Biochemistry, 37, 10047-61.

197

Konstantopoulou, I., Vassilopoulou, L., Mavragani-Tsipidou, P. and Scouras, Z.G.

(1992) Insecticidal effects of essential oils. A study of the effects of essential

oils extracted from eleven Greek aromatic plants on Drosophila auraria.

Experientia, 48, 616-9.

Lee, S., Tsao, R., Peterson, C. and Coats, J.R. (1997) Insecticidal activity of

monoterpenoids to western com rootworm (Coleoptera: Chrysomelidae),

twospotted spider mite (Acari: Tetranychidae), and house fly (Diptera:

Muscidae). Journal o f Economic Entomology, 90, 883-92.

Mabry, T.J. and Ulubelen, A. (1980) Chemistry and utilization of phenylpropanoids

including flavonoids, coumarins, and lignans. Journal o f Agricultural and

Food Chemistry, 28, 188-96.

Manoukas, A.G. and Mazomenos, B. (1977) Effect of antimicrobials upon eggs and

larvae of Dacus oleae (Diptera: Tephritidae) and the use of propionates for

larval diet preservation. Annales de Zoologie-Ecologie Animale, 9,277-85.

Matsumura, F. (1975) Toxicology o f Insecticides, Plenum Press, New York.

Matthews, G A. (1997) How to evaluate insecticide efficacy, in Methods in

Ecological and Agricultural Entomology (eds D.R. Dent and M.P. Walton),

CAB International, New York, pp. 243-69.

Mazomenos, B.E. and Haniotakis, G.E. (1981) A multicomponent female sex

pheromone of Dacus oleae Gmelin: Isolation and bioassay. Journal o f


Mazomenos, B.E. and Haniotakis, G.E. (1985) Male olive fruit fly attraction to

synthetic sex pheromone components in laboratory and field tests. Journal o f


Mazomenos, B.E., Pantazi-Mazomenos, A. and Stefanou, D. (2002) Attract and kill of

the olive fruit fly Bactrocera oleae in Greece as a part of an integrated control

system, in Use o f Pheromones and other Semiochemicals in Integrated

Control (eds P. Witzgall, B. Mazomenos and M. Konstantopoulou),

IOBC/WPRS Bulletin, 25,137-46.

Mazomenos, B.E., Stefanou, D., Langley, P. and Pantazi-Mazomenos, A. (1997)

Efeects of sugar-formulated triflumuron-treated targets on reproduction in the

198

olive fruit fly Bactrocera oleae (Diptera: Tephritidae). Bulletin o f

Entomological Research, 87, 169-72.

McChesney, J.D. (1994) After discovery: The issue of supply strategies in the

development of natural products, in Bioregulators fo r Crop Protection and

Pest Control (ed. P.A. Hedin), ACS Symposium Series No. 557, American


McCloud, E.S., Berenbaum, M.R. and Tuveson, R.W. (1992) Furanocoumarin content

and phototoxicity of rough lemon (Citrus jambhiri) foliage exposed to

enhanced ultraviolet-B (UVB) irradiation. Journal o f Chemical Ecology, 18,

1125-37.

McDonald, L.L., Guy, R.H. and Speirs, R.D. (1970) Preliminary evaluation of new

candidate materials as toxicants, repellents and attractants against stored-

product insects-I. Marketing Research Report No. 882, Agricultural Research

Service, United States Department o f Agriculture, Washington.

McLaren, J.S. (1986) Biologically active substances from higher plants: Status and

future potential. Pesticide Science, 17, 559-78.

McLaughlin, J.L., Zeng, L., Oberlies, N.H., Alfonso, D., Johnson, H.A. and

Cummings, B.A. (1997) Annonaceous acetogenins as new natural pesticides:

Recent progress, in Phytochemicals for Pest Control (eds P.A. Hedin, R.M.


Symposium Series No. 658, American Chemical Society, Washington, pp.

117-33.

Medina, P., Budia, F., del Estal, P. and Vifiuela, E. (2004) Influence of azadirachtin, a

botanical insecticide, on Chrysoperla carnea (Stephens) reproduction:

Toxicity and ultrastructural approach. Journal o f Economic Entomology, 97,

43-50.

Mendel, M.J., Alford, A.R. and Bentley, M.D. (1991a) A comparison of the effects of

limonin on Colorado potato beetle, Leptinotarsa decemlineata, and fall

armyworm, Spodoptera frugiperda, larval feeding. Entomologia

Experimental is et Applicata, 58, 191-4.

199

Mendel, M.J., Alford, A.R., Rajab, M.S. and Bentley, M.D. (1991b) Antifeedant

effects of Citrus limonoids differing in A-ring structure on Colorado potato

beetle (Coleoptera: Chrysomelidae) larvae. Journal o f Economic Entomology,

84,1158-62.

Mendel, M.J., Alford, A.R., Rajab, M.S. and Bentley, M.D. (1993) Relationship of

Citrus limonoid structure to feeding deterrence against fall armyworm

(Lepidoptera: Noctuidae) larvae. Environmental Entomology, 22, 167-73.

Miyakado, M., Watanabe, K. and Miyamoto, J. (1997) Natural products as leads in

structural modification studies yielding new agrochemicals, in Phytochemicals

for Pest Control (eds P.A. Hedin, R.M. Hollingworth, E.P. Masler, J.

Miyamoto and D.G. Thompson), ACS Symposium Series No. 658, American


Morgan, E.D. and Wilson, I.D. (1985) Methods and techniques for isolation of

pesticides, in CRC Handbook o f Natural Pesticides: Methods (ed. N.B.

Mandava), Vol. 2: Isolation and Identification, CRC Press, Boca Raton, pp. 3-

81.

Moyses, E.W. and Gfeller, F.J. (2001) Topical application as a method for comparing

the effectiveness of insecticides against cat flea (Siphonaptera: Pulicidae).

Journal o f Medical Entomology, 38, 193-6.

Murray, R.D.H (1991) Naturally occurring plant coumarins. Fortschritte der Chemie

organischer Naturstoffe, 58, 83-316.

Murray, R.D.H., Mendez, J. and Brown, S.A. (1982) The Natural Coumarins, John

Wiley & Sons, Chichester.

Musajo, L. and Rodighiero, G. (1962) The skin-photosensitizing furanocoumarins.

Experientia, 18, 153-200.

Nair, M.G. (1994) Natural products as sources of potential agrochemicals, in Natural

and Engineered Pest Management Agents (eds P.A. Hedin, J.J. Menn and

R.M. Hollingworth), ACS Symposium Series No. 551, American Chemical

Society, Washington, pp. 145-61.

Neal, J.J. (1989) Myristicin, saffole, and fagaramide as phytosynergists of

xanthotoxin. Journal o f Chemical Ecology, 15, 309-15.

2 0 0

Neal, J.J. and Wu, D. (1994) Inhibition of insect cytochromes P450 by

furanocoumarins. Pesticide Biochemistry and Physiology, 50,43-50.

Neuenschwander, P. (1982) Beneficial insects caught by yellow traps used in mass

trapping of the olive fly, Dacus oleae. Entomologia Experimental is et

Applicata, 32,286-96.

Nigg, H.N., Mallory, L.L., Fraser, S., Simpson, S.E., Robertson, J.L., Attaway, J.A.,

Callaham, S.B. and Brown, R.E. (1994) Test protocols and toxicity of

organophosphate insecticides to Caribbean fruit fly (Diptera: Tephritidae).

Journal o f Economic Entomology, 87, 589-95.

Nogata, Y., Yoza, K., Kusumoto, K., Kohyama, N., Sekiya, K. and Hideaki, O. (1996)

Screening for inhibitory activity of Citrus fruit extracts against platelet

cyclooxygenase and lipoxygenase. Journal o f Agricultural and Food

Chemistry, 44, 725-9.

Obeng-Ofori, D. and Reichmuth, C. (1997) Bioactivity of eugenol, a major

component of essential oil of Ocimum sauve (Wild.) against four species of

stored-product Coleoptera. International Journal o f Pest Management, 43, 89-

94.

Ohta, T., Maruyama, T., Nagahashi, M., Miyamoto, Y., Hosoi, S., Kiuchi, F.,

Yamazoe, Y. and Tsukamoto, S. (2002) Paradisin C: A new CYP3A4 inhibitor

from grapefruit juice. Tetrahedron, 58, 6631-5.

Patel, D. (1997) Liquid Chromatography, John Wiley & Sons, Chichester.

Peck, S.L. and Mcquate, G.T. (2000) Field tests of environmentally friendly

malathion replacements to suppress wild Mediterranean fruit fly (Diptera:

Tephritidae) populations. Journal o f Economic Entomology, 93, 280-9.

Pedigo, L.P. (1996) Entomology and Pest Management, 2nd edition, Prentice Hall,

London.

Plimmer, J.R. (1993) Regulatory problems associated with natural products and

biopesticides. Pesticide Science, 39, 103-8.

Poole, C.F. and Poole, S.K. (1991) Chromatography Today, Elsevier, Amsterdam.

2 0 1

Quayle, H.J. (1914) Citrus fruit insects in Mediterranean countries. United States

Department o f Agriculture Technical Bulletin, 134, 1-36.

Quayle, H.J. (1929) The Mediterranean and other fruit flies. University o f California

Experimental Station Circular, 315, 1-19.

Rani, P.U. and Osmani, F. (1980) Insecticidal, fumigant and repellent properties of

citral against house fly, Musca domestica nebulo (Fabr.). Pesticides, 8, 6-8.

Roark, R.C. (1942) The examination of plants for insecticidal constituents. Journal o f

Economic Entomology, 35, 273-6.

Robertson, J.L. and Preisler, H.K. (1992) Pesticide Bioassays with Arthropods, CRC

Press, Boca Raton.

Robertson, J.L., Smith, K.C., Savin, N.E. and Lavigne, R.J. (1984) Effects of dose

selection and sample size on the precision of lethal dose estimates in dose-

mortality regression. Journal o f Economic Entomology, 77, 833-7.

Rodriguez-Saona, C., Millar, J.G. and Trumble, J.T. (1997) Growth inhibitory,

insecticidal, and feeding deterrent effects of (12Z, 15Z)-l-acetoxy-2-hydroxy-

4-oxo-heneicosa-12,15-diene, a compound from avocado fruit, to Spodoptera

exigua. Journal o f Chemical Ecology, 23, 1819-31.

Rodriguez-Saona, C., Millar, J.G., Maynard, D.F. and Trumble, J.T. (1998) Novel

antifeedant and insecticidal compounds from avocado idioblast cell oil.

Journal o f Chemical Ecology, 24, 867-89.

Roelandts, R. (1984) Mutagenicity and carcinogenicity of methoxsalen plus UV-A.

Archives o f Dermatology, 120, 662-9.

Salvatore, A., Borkosky, S., Willink, E. and Bardon, A. (2004) Toxic effects of lemon

peel constituents on Ceratitis capitata. Journal o f Chemical Ecology, 30, 323-

33.

Sasaki, H., Taguchi, H., Endo, T. and Yosioka, I. (1980) The constituents of Glehnia

littoralis Fr. S ch m id t et M iq.. Structure of a new coumarin glycoside,

osthenol-7-0-/?-gentiobioside. Chemical & Pharmaceutical Bulletin, 28, 1847-

52.

2 0 2

Sastry, S.S., Ross, B.M. and Parraga, A. (1997) Cross-linking of DNA-binding

proteins to DNA with psoralen and psoralen furan-side monoadducts:

Comparison of action spectra with DNA-RNA cross-linking. Journal o f

Biological Chemistry, 272, 3715-23.

Schmutterer, H. (1990) Properties and potential of natural pesticides from the neem

tree, Azadirachta indica. Annual Reviews o f Entomology, 35, 271-97.

Schwarz, A.J., Liedo, J.P. and Hendrichs, J. (1989) Current programme in Mexico, in

Fruit Flies: Biology, Natural Enemies and Control, Vol. 3B (eds A. S.

Robinson and G. Hooper), Elsevier, Amsterdam, pp. 375-86.

Scott, B.R., Pathak, M.A. and Mohm, G.R. (1976) Molecular and genetic basis of

furanocoumarin reactions. Mutation Research, 39,29-74.

Scott, J.G., Ramaswamy, S.B., Matsumura, F. and Tanaka, K. (1986) Effect of

method of application on resistance to pyrethroid insecticides in Blattella

germanica (Orthoptera: Blattellidae). Journal o f Economic Entomology, 79,

571-5.

Serit, M., Ishida, M., Hagiwara, N., Kim, M., Yamamoto, T. and Takahashi, S. (1992)

Meliaceae and Rutaceae limonoids as termite antifeedants evaluated using

Reticulitermes speratus Kolbe (Isoptera: Rhinotermitidae). Journal o f


Sheppard, D.C. (1984) Toxicity of Citrus peel liquids to the house fly and red

imported fine ant. Journal o f Agricultural Entomology, 1, 95-100.

Silva, G.L., Lee, I. and Kinghom, A.D. (1998) Special problems with the extraction of

plants, in Natural Products Isolation (ed. R.J.P. Cannell), Methods in

Biotechnology 4, Humana Press, Totowa, pp. 343-63.

Siskos, E.P. (1997) Systemic effects of imidacloprid on the aphid Myzus persicae,

University of Bath, M.Sc. Thesis.

Stanley, W.L. and Jurd, L. (1971) Citrus coumarins. Journal o f Agricultural and Food

Chemistry, 19, 1106-10.

Stark, J.D. and Sherman, M. (1989) Toxicity, penetration and metabolism of acephate

in three fruit fly species (Diptera: Tephritidae). Journal o f Economic


203

Stark, J.D., Jepson, P.C. and Mayer, D.F. (1995) Limitations to use of topical toxicity

data for predictions of pesticide side effects in the field. Journal o f Economic


Stark, J.D., Vargas, R. and Miller, N. (2004) Toxicity of spinosad in protein bait to

three economically important tephritid fruit fly species (Diptera: Tepritidae)

and their parasitoids (Hymenoptera: Braconidae). Journal o f Economic


Stavroulakis, G., Adediran, K.A., Nikoloudi, A., Petrakis, C., Kalaitzaki, A. and

Michelakis, S. (2001) Rotenone: Efficiency against olive fly (Bactrocera oleae

Gmelin) and residual activity in olive oil. Biological Agriculture and

Horticulture, 19,207-17.

Stem, R.S., Thibodeau, L.A., Kleinerman, R.A., Parrish, J.A. and Fitzpatrick, T. B.

(1979) Risk of cutaneous carcinoma in patients treated with oral methoxsalen

photochemotherapy for psoriasis. New England Journal o f Medicine, 300,

809-13.-

Stevens, I.D.R. (1998) Chemical aspects of pheromones: Isolation and structure

determination, in Insect Pheromones and their Use in Pest Management (eds

P.E. Howse, I.D.R. Stevens and O.T. Jones), Chapman & Hall, London, pp.

180-225.

Stevenson, P.C., Simmonds, M.S.J., Yule, M.A., Veitch, N.C., Kite, G.C., Irwin, D.

and Legg, M. (2003) Insect antifeedant furanocoumarins from Tetradium

danielli. Phytochemistry, 63, 41-6.

Studebaker, G.E. and Kring, T.J. (2003) Effects of insecticides on Orius insidiosus

(Hemiptera: Anthocoridae), measured by field, greenhouse and Petri dish

bioassays. Florida Entomologist, 86, 178-85.

Styer, S.C. and Greany, P.D. (1983) Increased susceptibility of laboratory-reared vs.

wild Caribbean fruit fly, Anastrepha suspensa (Loew) (Diptera: Tephritidae),

larvae to toxic Citrus allelochemics. Environmental Entomology, 12, 1606-8.

Su, H.C.F. (1976) Toxicity of a chemical component of lemon oil to cowpea weevils.

Journal o f Georgia Entomological Society, 11,297-301.

204

Su, H.C.F. and Horvat, R. (1987) Isolation and characterization of four major

components from insecticidally active lemon peel extract. Journal o f

Agricultural and Food Chemistry, 35, 509-11.

Su, H.C.F., Speirs, R.D. and Mahany, P.G. (1972a) Citrus oils as protectants of black

eyed peas against cowpea weevils: Laboratory evaluation. Journal o f

Economic Entomology, 65, 1433-6.

Su, H.C.F., Speirs, R.D. and Mahany, P.G. (1972b) Toxicity of Citrus oils to several

stored-product insects: Laboratory evaluation. Journal o f Economic


Talukder, F.A. and Howse, P.E. (1994) Laboratory evaluation of toxic and repellent

properties of the pithraj tree, Aphanamixis polystachya Wall and Parker,

against Sitophilus oryzae (L.). International Journal o f Pest Management, 40,

274-9.

Taylor, W.E. and Vickery, B. (1974) Insectcidal properties of limonene, a constituent

of Citrus oil. Ghana Journal o f Agricultural Science, 7,61-2.

Thomas, C.J and Callaghan, A. (1999) The use of garlic (Allium sativa) and lemon

peel (Citrus limon) extracts as Culex pipiens larvacides: Persistence and

interaction with an organophosphate resistance mechanism. Chemosphere, 39,

2489-96.

Tirillini, B. (2000) Grapefruit: The last decade acquisitions. Fitoterapia, 71, S29-S37.

Touchstone, J.C. and Dobbins, M.F. (1983) Practice o f Thin Layer Chromatography,

2nd edition, John Wiley & Sons, New York.

Trumble, J.T., Dereks, W., Quiros, C.F. and Beier, R.C. (1990) Host plant resistance

and linear furanocoumarin content of Apium accessions. Journal o f Economic


Tsiropoulos, G.J. (1972) Storage temperatures for eggs and pupae of the olive fruit

fly. Journal o f Economic Entomology, 65, 100-2.

Tsiropoulos, G.J. and Manoukas, A.G. (1977) Adult quality of Dacus oleae affected

by larval crowding and pupal irradiation. Annals o f the Entomological Society

o f America, 70, 916-8.

205

Tsitsipis, J.A. (1975) Mass rearing of the olive fruit fly, Dacus oleae (Gmelin), at

‘Demokritos’, in Controlling Fruit Flies by the Sterile-Insect Technique,

International Atomic Energy Agency, Vienna, STI/PUB/392, pp. 93-100.

Tsitsipis, J.A. (1977a) Development of a caging and egging system for mass rearing

the olive fruit fly, Dacus oleae (Gmel.) (Diptera: Tephritidae). Zeitschrift fur

Angewandte Entomologie, 83, 96-105.

Tsitsipis, J.A. (1977b) An improved method for the mass rearing of the olive fruit fly,

Dacus oleae (Gmel.) (Diptera: Tephritidae). Zeitschrift fu r Angewandte

Entomologie, 83, 419-26.

Tukey, J.W. (1953) Some selected quick and easy methods of statistical analysis.

Transactions o f the New York Academy o f Sciences, Series 2, 16, 88-97.

Velenciennes, E., Smadja, J. and Conan, J.Y. (1999) Screening for biological activity

and chemical composition of Euodia borbonica var. borbonica (Rutaceae), a

medicinal plant in Reunion Island. Journal o f Ethnopharmacology, 64,283-8.

Wagstaff, D.J. (1991) Dietary exposure to furanocoumarins. Regulatory Toxicology

and Pharmacology, 14, 261-72.

Walton, M.P. (1995) Integrated pest management in olives, in Integrated Pest

Management (ed. D. Dent), Chapman & Hall, London, pp. 222-40.

Wangensteen, H., Molden, E., Christensen, H. and Malterud, K.E. (2003)

Identification of epoxybergamottin as a CYP3A4 inhibitor in grapefruit peel.

European Journal o f Clinical Pharmacology, 58, 663-8.

Weaver, D.K., Wells, C.D., Dunkel, F.V., Bertsch, W., Sing, S.E. and Sriharan, S.

(1994) Insecticidal activity of floral, foliar and root extracts of Tagetes minuta

(Asterales: Asteraceae) against adult Mexican bean weevils (Coleoptera:

Bruchidae). Journal o f Economic Entomology, 87, 1718-25.

Weaver, D.K., Zettler, L.J., Wells, C.D., Baker, J.E., Bertsch, W. and Throne, J.E.

(1997) Toxicity o f fractionated and degraded Mexican marigold floral extract

to adult Sitophilus zeamais (Coleoptera: Curculionidae). Journal o f Economic


Williams, D.H. and Fleming, I. (1995) Spectroscopic Methods in Organic Chemistry,

5th edition, McGraw-Hill, London.

206

Wong, T.T. Y., Ramadan, M.M., Herr, J.C. and Mcinnis, D.O. (1992) Suppression of a

Mediterranean fruit fly (Diptera: Tephritidae) population with concurrent

parasitoid and sterile fly releases in Kula, Maui, Hawaii. Journal o f Economic


Yang, L.L., Wang, M.C., Chen, L.G. and Wang, C.C. (2003) Cytotoxic activity of

coumarins from the fruits of Cnidium monnieri on leukaemia cell lines. Planta

Medica, 62, 1091-5.

Zobel, A.M., Wang, J., March, R.E. and Brown, S.A. (1991) Identification of eight

coumarins occurring with psoralen, xanthotoxin, and bergapten on leaf

surfaces. Journal o f Chemical Ecology, 17, 1859-70.

207

Documents

INSECTICIDAL ACTIONS OF CITRUS AURANTIUM